Vaithilingam, Jayasheelan (2015) Additive manufacturing and...

Vaithilingam, Jayasheelan (2015) Additive manufacturing and surface functionalisation of Ti6Al4V components using self-assembled monolayers for biomedical applications. PhD thesis, University of Nottingham.

Access from the University of Nottingham repository: http://eprints.nottingham.ac.uk/28474/1/Jayasheelan_Vaithilingam_PhD_Thesis_2015.pdf

Copyright and reuse:

The Nottingham ePrints service makes this work by researchers of the University of Nottingham available open access under the following conditions.

This article is made available under the University of Nottingham End User licence and may be reused according to the conditions of the licence. For more details see: http://eprints.nottingham.ac.uk/end_user_agreement.pdf

For more information, please contact [email protected]

mailto:[email protected]

ADDITIVE MANUFACTURING AND SURFACE

FUNCTIONALISATION OF Ti6Al4V

COMPONENTS USING SELF-ASSEMBLED

MONOLAYERS FOR BIOMEDICAL

APPLICATIONS

Jayasheelan Vaithilingam, B.Tech., MSc.

Thesis submitted to the University of Nottingham for the degree of Doctor of

Philosophy

July 2015

Abstract i

Abstract

The ability to provide mass customised and biocompatible implants is

increasingly important to improve the quality of life. Additive manufacturing

(AM) techniques have obtained increasing popularity and selective laser

melting (SLM), a metal-based AM technique with an ability to build complex

and well defined porous structures, has been identified as a route to fabricate

customised biomedical implants. Surface modification of an implant with a

biomolecule is used to improve its biocompatibility and to reduce post-implant

complications. In this thesis, the potential of a novel approach to use self-

assembled monolayers to modify SLM fabricated surfaces with therapeutic

drugs has been evaluated.

Although there are numerous studies on the material development, process

optimisation and mechanical testing of SLM fabricated parts, the surface

chemistry of these parts is poorly understood. Initially, the surface chemistry

of SLM as-fabricated (SLM-AF), SLM fabricated and mechanically polished

(SLM-MP) and forged and mechanically polished (FGD-MP) parts made of

Ti6Al4V was determined using an X-ray photoelectron spectrophotometer

(XPS). Later the impact of laser power on the surface chemistry of the parts

was also studied. A non-homogeneous surface chemistry was observed due to

a change in the distribution of the alloying elements titanium, aluminium and

vanadium on the surface oxide layer. Surface modification of the SLM

fabricated component would be beneficial to obtain a homogenous surface

chemistry, especially for biomedical application.

Coating of self-assembled monolayers (SAMs) onto SLM fabricated Ti6Al4V

structures was performed to modify their surface chemistry. 16-

phosphanohexadecanoic acid monolayers (16-PhDA) were used to modify

SLM-AF and SLM-MP surfaces. XPS and static water contact angle

measurements confirmed the chemisorption of monolayers on these surfaces.

The obtained results confirmed that SAMs were stable on the Ti6Al4V surface

for over 28 days before its desorption. It was also witnessed that the stability of

monolayers on the rough SLM-AF and smooth SLM-MP surfaces were not

Abstract ii

significantly different. Later, the 16-PhDA SAM coated Ti6Al4V SLM-MP

surface was functionalised with a model drug, Paracetamol. An esterification

reaction was performed to functionalise the phosphonic acid monolayers with

Paracetamol. Surface characterisation revealed the sucessful attachment of

Paracetamol to the SAMs.

Bacterial infections from biomedical implants and surgical devices are reported

to be a major problem in orthopaedic, dental and vascular surgery. Hence, to

further explore the potential of the proposed method, Ciprofloxacin® a broad

spectrum antibiotic was immobilised to the SAMs, previously adsorbed on the

SLM-MP Ti6Al4V surfaces. Using the proposed approach, approximately 1.12

µg/cm2 of the drug was coated to the surface. Results showed that

Ciprofloxacin® is highly stable under the oxidative conditions used in this

study. Under in vitro condition, the drug was observed to release in a sustained

manner. Antibacterial susceptibility tests revealed that the immobilised

Ciprofloxacin® was therapeutically active upon its release. Thus, a novel

methodology to fabricate customised and functionalised implants has been

demonstrated for an improved biocompatibility and reduced post-implant

complications.

List of Publications iii

List of Publications

Journals Published

1. Reika Sakai, Baiju John, Masami Okamoto, Jukka V. Seppälä,

Jayasheelan Vaithilingam, Husnah Hussein and Ruth Goodridge

(2012). Fabrication of poly-lactide based biodegradable thermoset

scaffolds for tissue engineering applications. Macromolecular and

Materials Engineering; 298, 45-52.

2. Jayasheelan Vaithilingam, Ruth D. Goodridge, Steven D.R. Christie,

Steve Edmondson and Richard J.M. Hague (2014). Immobilisation of

antibacterial drug – Ciprofloxacin® to Ti6Al4V components fabricated

using selective laser melting. Applied Surface Science; 314, 642-654.


Steve Edmondson and Richard J.M. Hague (2015). Functionalisation of

Ti6Al4V components fabricated using selective laser melting with a

biomolecule. Materials Science and Engineering: C; 46, 52-61.

Book Chapter Published


Steve Edmondson and Richard J.M. Hague (2015). Additive

manufacturing and surface modification of biomaterials using self-

assembled monolayers. In: Dietmar W. Hutmacher and Wojciech

Chrzanowski (eds), Biointerfaces: Where Material Meets Biology First

Edition; Royal Society of Chemistry. 30-54 (ISBN: 9781849738767).

Manuscript Submitted


Steve Edmondson and Richard J.M. Hague. An investigation on the

surface chemistry of selective laser melted and forged Ti6Al4V

components. (Advanced Materials Engineering).

List of Publications iv

Manuscript In-preparation


Steve Edmondson and Richard J.M. Hague. The effect of laser re-

melting on the surface chemistry of Ti6Al4V components fabricated by

selective laser melting. (To be submitted to Surface Science).


Steve Edmondson and Richard J.M. Hague. Aging effect of Ti6Al4V

powders during selective laser melting process. (To be submitted to

Journal of Materials Processing Technology).


Steve Edmondson and Richard J.M. Hague. Electropolishing of

tracheobronchial stents fabricated by selective laser melting. (To be

submitted to Biomedical Materials Research Part B: Applied

Biomaterials).

Selected Conference Presentations

List of Selected Oral Presentations


Steve Edmondson and Richard J.M. Hague, 2013. Anti-bacterial

activity of Ciprofloxacin® immobilised to self-assembled monolayers;

25th

European Society for Biomaterials Annual Conference, Madrid,

Spain.


Steve Edmondson and Richard J.M. Hague, 2013. Delivery of an anti-

bacterial drug from an additively manufactured surface; 24th

International Conference on Solid Free-form Fabrication, Texas, USA.

3. Jayasheelan Vaithilingam, Ruth D Goodridge, Steven D Christie, Steve

Edmondson and Richard JM Hague, 2012. Surface modification of

selective laser melted structures using self-assembled monolayers for

biomedical applications; 23rd

Inter-national Conference on Solid

Freeform Fabrication, Texas, USA. 316-325.

4. Jayasheelan Vaithilingam, Ruth D. Goodridge, Richard J.M. Hague,

Steven D.R. Christie, and Steve Edmondson, 2012. Fabrication of

List of Publications v

customised biomedical implants; 2nd

Annual Research Student

Conference, Health and Life Sciences Research School, Loughborough

University, UK.


Steven D.R. Christie, and Steve Edmondson, 2012. Fabrication and

surface modification of customised biomedical implants; In: East

Midlands Universities Postgraduate Research Students Conference,

University of Nottingham, Sutton Bonington, UK.


Steven D. Christie, and Steve Edmondson, 2012. Surface modification

of selective laser melted Ti6Al4V structures using self-assembled

monolayers; In: The UK Society for Biomaterials Annual Conference,

University of Nottingham, UK.

7. Jayasheelan Vaithilingam, Ruth D. Goodridge, Richard J.M. Hague and

Steven D Christie, 2012. Additive manufacturing and surface

modification of tracheobronchial stents using self-assembled

monolayers; 9th

World Biomaterials Conference, Chengdu, China.

List of Selected Poster Presentations


Richard J.M. Hague and Steve Edmondson, 2013. Immobilisation of

Ciprofloxacin® to an additively manufactured Ti6Al4V surface. 25th

European Society for Biomaterials Annual Conference, Madrid, Spain.


Steven D.R. Christie, and Steve Edmondson, 2013. Revolutionising the

fabrication of biomedical implants; Research Showcase Event,

University of Nottingham, UK.


Steven D.R. Christie, and Steve Edmondson, 2013. A novel approach

to engineer customised and functionalised biomedical implants; In:

Engineering Postgraduate Research Event, Faculty of Engineering,

University of Nottingham, UK (Finalist).


Steven D.R. Christie, and Steve Edmondson, 2013. A novel approach

List of Publications vi

to engineer customised and functionalised biomedical implants; In:

Science, Engineering and Technology for Britain (SET for Britain),

The House of Commons, Westminster, London, UK.


Steven D.R. Christie, and Steve Edmondson, 2012. Fabrication and

functionalisation of customised biomedical implants; In:

Loughborough Research That Matters: Annual Research Conference,

Loughborough University, UK.

6. Jayasheelan Vaithilingam, Ruth D Goodridge, Richard J.M. Hague,

Steven D.R. Christie, and Steve Edmondson, 2011. Fabrication of

airway stents by selective laser melting; 1st Annual Research Student

Conference, Health and Life Sciences Research School, Loughborough

University, UK.

Acknowledgement vii

Acknowledgements

Right from the start of my research, I came across several great people and I

strongly believe it would not be possible for me to progress to this stage of my

research without the assistance and mentorship of such great people and here

are few of them.

I am indebted to my supervisors Dr. Ruth Goodridge and Prof. Richard Hague,

Additive Manufacturing and 3D Printing Research Group (AM & 3DPRG),

Department of Mechanical, Materials and Manufacturing Engineering,

University of Nottingham for their dedicated supervision, priceless guidance

and advice during my research period. Technical assistance rendered by my

supervisors kept me updated with the recent developments on mechanical &

material processing. Without their continued support, motivation and

assistance it would have been difficult for me to complete my PhD.

A special thanks to my supervisor Dr. Ruth Goodridge for her suggestions and

helping me with research papers and thesis writing. Her motivations and

understandings helped me to a lot during the course.

I am very thankful to Dr. Steven Christie, Department of Chemistry,

Loughborough University for his co-supervision, constant technical support

and motivation during my experimentations at the Department of Chemistry,

Loughborough University. I also owe gratitude to Dr. Steve Edmondson,

School of Materials, University of Manchester, for his co-supervision,

guidance and technical assistance on surface characterisation and surface

modification techniques. I am grateful to Dr. Chris Tuck, Prof. Ricky Wildman

and Prof. David Grant, University of Nottingham for their valuable suggestions

and assisting with my yearly assessments. I thank Prof. Wayne Hayes,

University of Reading for assessing my thesis and for his valuable suggestions

during my PhD viva.

I also would like to thank AM & 3DPRG technicians Mr. Mark East and Mr.

Mark Hardy for their technical support and immense help during my

experimental works “Rapids Bro”. I extend my humble thanks to the

Acknowledgement viii

technicians, Mr. Andy Sandaver, Mr. Jagpal Singh, Mrs. Pat Cropper, Dr.

Keith Yendall and Mrs. Marion Dillon at Loughborough University for their

technical support and wise suggestions during my experimentations.

I would like to thank all my colleagues and ex-colleagues of AM & 3DPRG

members in particular Dave, Ajit, James, Helen, Martin, Mirela, Adedeji,

Momo, Jiaming, Farhan, Marco, Belen, Joe, Chandra and Meisam for their

help and assistance. I am also thankful to Sam, Andy, James, Tom and Nuria at

the Department of Chemistry, Loughborough University for their help and

support during my experimentations.

I also thank Prof. Nat Puri (Puri Foundation), Mr. Vydeswaran Murugesan, Dr.

Bindhumadhavan Gururajan, Dr. Ayarivan Puratchikody, Prof. Zoltan Nagy,

Dr. Aruna Palipana and Mr. Ramesh Kumar Muthusamy for their motivation,

support and guidance.

I express my sincere thanks to my uncles Mr. Subramani Chidambaram, Mr.

Palanivelu Chidambaram, Mr. Baskar Chidambaram, my aunties Mrs. Chitra

Sivashanmugam, Mrs. Amsavalli Manohar and their family for all their help,

support, care and everything they did for me.

I am grateful to my friends Sarvi, Lalitha, Alsu, Dipesh, Sushma, Shridharan,

Pradeep, Sundar, Subash, Deepak, Ramji, Matt, Darren, Alun, Dave, Sonal,

Sarah, Sophie, Andrea, John, Jodie and Jean for their help and support

throughout my PhD.

I express my tremendous gratitude to my best friends Sabari, Manikandan,

Ranjani, Santosh, Sanjeevi, Jayaraj, Dinakar Babu, Bharathi and Alex for all

their help and support to me and my family. Friends like you are really hard to

find.

Finally, I would like to thank the University of Nottingham, Loughborough

University and Engineering and Physical Sciences Research Council (EPSRC),

UK for funding my research.

Dedication ix

Dedication

This thesis is dedicated to:

My mum, Jaya, my beloved dad, Mr. Vaithilingam Vaiyapuri and my paternal

grand mum, Mrs. Sellammaal Vaiyapuri for their myriad sacrifices throughout

their life for me. I am so proud and happy that I had you both in my life.

Although you are not with me now physically, I know you both truly lived one

and only for me. I would never say good bye to you cos, I know you will

always be there in and around me. Your love, humbleness, kindness and your

other numerous qualities instilled me and will reflect in each of my

endeavours. Dear dad and grandma, “I MISS YOU” is not even the right word

to express how I actually feel.

Mrs. Mariyayee Chidambaram, my maternal grand “mum” for her immense

effort in bringing me up. I wouldn’t be here in this position without all your

sacrifices. I could never ask for a better grand mum than you. I LOVE YOU.

My teachers and supervisors (undergraduate, masters and PhD) for their huge

support, guidance and advice throughout my life till now. Every one of you

have given me the most precious time in my life that I cannot claim back

anymore. I have learnt a lot of good qualities from you all and I certainly

believe they will greatly profile my career.

Mr. Muruganantham Kanagasabai, Mr. Natarajan Subramanian, Dr. Ayarivan

Puratchikody and Dr. Rajendran Ramasamy for fostering scientific temper in

me and for their continued encouragement and support in all my endeavours.

My friend, Mr. Anbarasan Subramanian. Although I envisioned doing a PhD

from my childhood, you are an important person who kindled my aim further.

My sisters Hemalatha Rajkumar, Deepa Vijayakumar and Sharmila Paul

Reuben and their family who have given me the unconditional love and

support in every possible way they could. I am very fortunate to have you all.

My wife and friend for life, Mrs. Ramya Jayasheelan. Thank you for your

motivation, care, love and everything that you have given me. I LOVE YOU!

Table of Contents x

CONTENTS Abstract ................................................................................................................ i

List of Publications ........................................................................................... iii

Acknowledgements...........................................................................................vii

Dedication .......................................................................................................... ix

List of Figures .................................................................................................. xiv

List of Schemes................................................................................................ xxi

List of Tables ..................................................................................................xxii

List of Symbols and Units ............................................................................. xxiv

List of Acronyms ............................................................................................ xxv

1. INTRODUCTION ..................................................................................... 1

1.1. Biomaterials ......................................................................................... 1

1.2. Market for Biomedical Implants .......................................................... 3

1.3. Biological Response to an Implant ...................................................... 4

1.4. Issues with Current Biomedical Implants ............................................ 6

1.4.1. Customisation and fabrication of complex geometries ................ 6

1.4.2. Post-implant complications .......................................................... 8

1.5. Current Research ................................................................................ 12

1.6. Scope of the Research ........................................................................ 13

1.7. Thesis Structure ................................................................................. 14

2. LITERATURE REVIEW ....................................................................... 15

2.1. Introduction ........................................................................................ 15

2.2. Additive Manufacturing ..................................................................... 15

2.3. Selective Laser Melting ..................................................................... 17

2.3.1. Generation of a CAD file and supports for SLM ....................... 17

2.3.2. Equipment and Principle of SLM ............................................... 18

2.3.3. Factors governing the SLM process ........................................... 19

2.3.4. Heat Transfer in a SLM Process ................................................. 24

2.3.5. Melt Pool Dynamics ................................................................... 25

2.3.6. Microstructural Development and Thermal Effects ................... 26

2.3.7. Existing research on SLM .......................................................... 27

2.3.8. Fabrication of medical devices using SLM ................................ 33

2.4. Surface Modification ......................................................................... 37

Table of Contents xi

2.4.1. Issues with current surface modification techniques .................. 38

2.5. Surface Modification Using Self-Assembled Monolayers ................ 39

2.5.1. Formation of SAMs .................................................................... 40

2.5.2. Assembly of SAMs by simple immersion .................................. 41

2.5.3. Types of SAMs ........................................................................... 43

2.5.4. Biomedical Applications of Phosphonate SAMs ....................... 47

2.6. Knowledge Gap ................................................................................. 51

2.6.1. SLM ............................................................................................ 51

2.6.2. SAMs .......................................................................................... 53

2.7. Summary ............................................................................................ 54

3. RESEARCH HYPOTHESIS AND NOVELTY ................................... 55

3.1. Aim of the Research ........................................................................... 55

3.2. Objectives .......................................................................................... 55

3.3. Research Novelty and Contribution to Existing Knowledge ............. 55

4. RESEARCH METHODOLOGY ........................................................... 57

4.1. Summary of Experimental Procedure ................................................ 57

4.1.1. Materials ..................................................................................... 60

4.1.2. Equipment ................................................................................... 61

4.1.3. Methods ...................................................................................... 66

5. SURFACE MORPHOLOGY AND SURFACE CHEMISTRY .......... 79

5.1. Introduction ........................................................................................ 79

5.2. Effect of Aging of Ti6Al4V Powders ................................................ 79

5.2.1. Surface Morphology ................................................................... 79

5.2.2. Particle Size Distribution ............................................................ 81

5.2.3. Surface Chemistry ...................................................................... 83

5.3. SLM Surface Characterisation ........................................................... 87

5.3.1. Surface Morphology and Roughness .......................................... 87

5.3.2. Surface Chemistry ...................................................................... 89

5.4. Surface Chemistry Comparisons ...................................................... 100

5.4.1. Effect of skin scanning ............................................................. 100

5.4.2. Top Surface and Side Surface................................................... 107

5.4.3. Recycled powders Vs side surface ........................................... 108

5.5. Summary .......................................................................................... 110

Table of Contents xii

6. SURFACE MODIFICATION USING SAMs ..................................... 112

6.1. Introduction ...................................................................................... 112

6.2. SAM Attachment ............................................................................. 112

6.2.1. Surface chemistry ..................................................................... 113

6.2.2. Surface wettability .................................................................... 114

6.3. Stability Studies ............................................................................... 116



6.4. Summary .......................................................................................... 118

7. FUNCTIONALISATION OF SAMs ................................................... 120

7.1. Introduction ...................................................................................... 120

7.2. SAM attachment .............................................................................. 120



7.3. SAMs Functionalisation with Paracetamol ...................................... 123



7.4. Summary .......................................................................................... 132

8. IMMOBILISATION OF CIPROFLOXACIN® ................................ 134

8.1. Introduction ...................................................................................... 134

8.2. SAM attachment .............................................................................. 134



8.3. Functionalisation of SAMs with Ciprofloxacin® ............................ 138



8.4. Stability Studies ............................................................................... 145

8.4.1. Oxidative stability of Ciprofloxacin® ...................................... 145

8.4.2. In vitro stability of Ciprofloxacin® .......................................... 147

8.5. Drug Quantification ......................................................................... 151

8.6. Drug Elution Studies ........................................................................ 151

8.7. Antibacterial Susceptibility .............................................................. 153

8.7.1. Test-1 ........................................................................................ 153

Table of Contents xiii

8.7.2. Test-2 ........................................................................................ 156

8.8. Summary .......................................................................................... 157

9. CONCLUSION AND FUTURE WORK ............................................. 158

9.1. Conclusion ....................................................................................... 158

9.2. Future Work ..................................................................................... 159

REFERENCES ............................................................................................. 161

APPENDIX 1 ................................................................................................. 178

APPENDIX 2 ................................................................................................. 196

APPENDIX 3 ................................................................................................. 198

List of Figures xiv

List of Figures

Figure 1-1 Biomedical implants for human application. Reproduced with kind

permission from Bentham Open [3]. ........................................................... 2

Figure 1-2 Schematic of the sequential reactions that take place after the

implantation of a biomaterial. Figure reproduced with kind permission

from Elsevier [4]. ........................................................................................ 5

Figure 2-1 Schematic of the SLM process. Reproduced with kind permission

from Vision Systems [80].......................................................................... 19

Figure 2-2 Schematic representation of laser scanning related input process

parameters (a); SLM fabricated Ti6Al4V part showing actual laser scan

track (b) and further magnified area depicting point distance of the laser

scan(c). ...................................................................................................... 22

Figure 2-3 Schematic of some of the scan strategies typically used in a SLM

process. Unidirectional (a); bi-directional (b) and alternating bi-directional

(c)............................................................................................................... 23

Figure 2-4 Schematic representation of melt pool (a) and splitting of melt pool

resulting in balling phenomena (b). ........................................................... 26

Figure 2-5 Micrograph showing a polished 316L SS surface depicting a

balling induced pore. Reproduced with kind permission from Elsevier

[68]. ........................................................................................................... 30

Figure 2-6 Typical surface morphology of a part fabricated using a Renishaw

AM 250 machine. Note that the SEM micrograph is tilted to

approximately 45° to the right to its build direction. ................................ 35

Figure 2-7 Customised cranial implant fabricated using the SLM process.

SLM fabricated implant (a); Implant with support structures (b); Implant

removed from support structure showing inner geometry (c) and custom-

fitting of the implant to the skull (d). [Implant design and 3D model

courtesy of CARTIS]................................................................................. 36

Figure 2-8 Customised maxillo facial implant fabricated using SLM. CAD

design of a customised maxillo-facial implant (a); Implants fabricated in a

SLM machine (b); Implant with its support structure (c); Custom fitting of

List of Figures xv

implant to a model of the damaged part (d) [Implant design and 3D model

courtesy of CARTIS]................................................................................. 36

Figure 2-9 Micrographs of BiodivYsio (a, b) Taxus (c, d) and Cypher (e, f)

stents commercially available for cardiovascular stenting. The stents

exhibit cracking and peeling of the polymeric coatings on their expansion.

Reproduced with kind permission from HMP Communications [44]. ..... 39

Figure 2-10 Schematic diagram of a single crystalline alkanethiol SAM

formed on a metal surface. ........................................................................ 39

Figure 2-11 Steps involved in the adsorption (a) and assembly (b) of SAMs. 42

Figure 2-12 Schematic representation of the binding mechanism of

phosphonic acid SAMs to a metal surface. ............................................... 46

Figure 2-13 Atomic force microscope images of flufenamic acid

functionalised on gold surface using acid chloride esterification (a) and

flufenamic acid coated to a titanium surface by dry heat esterification

method (b). Reproduced with kind permission from Elsevier [37]. .......... 50

Figure 4-1 Schematic of the research methodology ........................................ 59

Figure 4-2 Spectrum 100 Fourier Transform Infrared spectrometer fitted with

a pressure arm (with the kind permission from Perkin Elmer). ................ 64

Figure 4-3 Schematic of a UV-Visual Spectrophotometer .............................. 65

Figure 4-4 Representation of sample models aligned in the build platform and

the Meander scan strategy used for fabrication (right hand corner) .......... 67

Figure 4-5 Maximum absorbance observed for Ciprofloxacin® ..................... 74

Figure 4-6 Calibration Graph for UV-Vis spectrophotometer to determine the

unknown concentration for the drug quantification experiment. .............. 74

Figure 4-7 Calibration Graph for UV-Vis spectrophotometer to determine

unknown concentration for the drug elution experiment. ......................... 75

Figure 4-8 Schematic representation of test specimens placed on a culture

plate ........................................................................................................... 77

Figure 4-9 Schematic representation of test specimens placed on a culture

plate. Label description: C- control disc with no drug; CD – control with

drug on; TC: Control disc coated with Tris-HCl buffer; 1, 2, 4, 6 –

immersion time intervals (in weeks) of the samples in buffer solution. ... 78

Figure 5-1 Surface morphology of virgin (a and b) and recycled (c and d)

Ti6Al4V particles obtained using SEM at different magnifications. ........ 80

List of Figures xvi

Figure 5-2 SEM micrographs of Ti6Al4V recycled powders retained after

sieving using a sieve of diameter of 63 µm. .............................................. 81

Figure 5-3 Comparison of particle size distribution obtained for virgin,

recycled and sieve retained Ti6Al4V particles. ......................................... 81

Figure 5-4 Survey Spectra obtained for virgin and recycled Ti6Al4V powders

using XPS. ................................................................................................. 83

Figure 5-5 High resolution spectra of C 1s, O 1s and N 1s regions obtained for

virgin and recycled Ti6Al4V powders using XPS. .................................. 84

Figure 5-6 Comparison of high resolution spectra of C 1s, O 1s and N 1s

regions obtained for virgin and recycled Ti6Al4V powders using XPS.

The dotted lines show a possible change in the elemental composition of

recycled powders due to contamination from the atmosphere. ................. 85

Figure 5-7 High resolution spectra of Ti 2p, Al 2p and V 2p regions obtained

for virgin and recycled Ti6Al4V powders using XPS. ............................. 86

Figure 5-8 SEM images showing the surface morphology of an as-fabricated

Ti6Al4V surface fabricated using SLM. The figure clearly depicts the

rough and porous nature of the SLM-AF surface...................................... 88

Figure 5-9 Surface Topography of an SLM as-fabricated Ti6Al4V surface

obtained using a surface profilometer. The measured surface area was 175

µm2. ........................................................................................................... 88

Figure 5-10 Survey spectra of SLM as-fabricated (SLM-AF), SLM

mechanically polished (SLM-MP) and forged mechanically polished

(FGD-MP) Ti6Al4V surface. .................................................................... 89

Figure 5-11 Depth profile spectra of C 1s and O 1s regions for SLM-AF,

SLM-MP and FGD-MP samples. Depth profiling was performed every 30

seconds (~ 4 nm) from 0 to 270 seconds. C 1s spectra show the

disappearance of the initially observed peak after 30 seconds of etching

which is attributed to surface contamination. Reduction in the intensity of

O 1s peak on increasing depth can also be witnessed from the figure. ..... 91

Figure 5-12 High resolution XPS scan of Ti 2p region showing the oxidation

states of titanium on a Ti6Al4V SLM-AF surface. ................................... 94

Figure 5-13 Depth profiles obtained for titanium 2p region using XPS for

SLM-AF, SLM-MP and FGD-MP surfaces. Depth profiling was

List of Figures xvii

performed at every 30 seconds (approximately 4 nm) from 0 to 270

seconds. ..................................................................................................... 94

Figure 5-14 Depth profiles obtained for aluminium 2p region using XPS for



seconds. ..................................................................................................... 95

Figure 5-15 Depth profiles obtained for vanadium 2p region using XPS for



seconds. ..................................................................................................... 96

Figure 5-16 Estimation of oxide later thickness for SLM-AF, SLM-MP and

FGD-MP samples using two different methods. Data points at every 40

angstroms................................................................................................... 97

Figure 5-17 Schematic of XPS probing on a rough SLM-AF Ti6Al4V

component. ................................................................................................ 98

Figure 5-18 Surface morphology of non-skin scanned and skin scanned

Ti6Al4V SLM surfaces depicting the laser scan tracts. .......................... 101

Figure 5-19 Surface morphology of a non-skin scanned (NSK) (a) and skin

scanned (SK) (b) Ti6Al4V part fabricated by SLM. ............................... 101

Figure 5-20 XPS spectra of C 1s, N 1s and O 1s regions for the non-skin

scanned (NSK) and skin scanned (SK) Ti6Al4V surfaces fabricated by

SLM. The dotted lines shows the peaks that disappeared on etching the

surface. .................................................................................................... 103

Figure 5-21 XPS spectra of Ti 2p, Al 2p and V 2p regions for the non-skin

scanned (NSK) and skin scanned (SK) Ti6Al4V surfaces fabricated by

SLM. The dotted lines show the transition of metal oxides from the

outermost layer to pure metals. ............................................................... 104

Figure 5-22 Evolution of the surface chemistry of non-skin scanned (a) and

skin scanned (b) SLM fabricated surface. ............................................... 106

Figure 5-23 SEM micrographs of a side surface (a) and top surface (b). ...... 107

Figure 5-24 Surface morphology of recycled powders used to fabricate the

part (a) and the side surface with partially sintered particles (b). ........... 109

Figure 5-25 Elemental ratio of titanium and aluminium to oxygen in the

recycled powder and the side surface. ..................................................... 110

List of Figures xviii

Figure 6-1 XPS spectra before and after surface modification of a) SLM-AF

and b) SLM-MP by adsorption of a 16-PhDA monolayer. ..................... 114

Figure 6-2 Static water contact angle measurements on SLM-AF and SLM-

MP surfaces, after cleaning and SAM attachment. ................................. 115

Figure 6-3 XPS spectra for the in vitro stability of SAMs on SLM-AF (a) and

SLM-MP (b) surfaces after immersing in Tris-HCl buffer solution for

various time intervals. ............................................................................. 117

Figure 6-4 Static water contact angle measurements on SLM as-fabricated

(SLM-AF) and SLM-MP surfaces for different immersion time intervals

in Tris-HCl buffer solution. “Before SM” refers to samples before surface

modification using SAMs and “Week 0” measurements made after surface

modification............................................................................................. 118

Figure 7-1 Survey spectra showing the change in surface chemistry of a

Ti6Al4V surface after SAM attachment. (a) Control surface without SAM

coating and (b) After 16-PhDA SAM coating......................................... 121

Figure 7-2 Survey spectra obtained using XPS for the SAM coated and

Paracetamol coated surfaces. The changes in the intensity of the detected

elements before and after drug coating indicate a change in the surface

chemistry. ................................................................................................ 124

Figure 7-3 Ratio of carbon and oxygen to its underlying metals and C/O ratio

for control, SAM coated and Paracetamol coated surfaces. .................... 126

Figure 7-4 High resolution spectra of carbon 1s region obtained for (a)

control, (b) SAM attached, (c) Paracetamol coated samples, (d) shows the

C 1s region for Paracetamol powder characterised using XPS. .............. 127

Figure 7-5 XPS spectra obtained for the oxygen 1s region for (a) control, (b)

SAM attached, (c) Paracetamol coated samples, (d) shows the O 1s region

for Paracetamol powder characterised using XPS. ................................. 129

Figure 7-6 N 1s region for (a) control, (b) SAM attached, (c) Paracetamol

coated samples, (d) Paracetamol powder, characterised using XPS. ...... 130

Figure 7-7 Contact angles obtained for SLM fabricated Ti6Al4V surfaces after

cleaning (a), 16-PhDA SAM attachment (b) and Paracetamol attachment

(c)............................................................................................................. 131

Figure 7-8 Static water contact angle obtained for Control, SAM coated and

Paracetamol coated Ti6Al4V samples. ................................................... 131

List of Figures xix

Figure 8-1 Survey spectra shows the change in the intensity of C 1s peak

before and after SAM attachment and the introduction of P 2p peak after

SAM attachment thus confirming the attachment of 16-PhDA SAMs and

ODPA SAMs to the Ti6Al4V surface. .................................................... 134

Figure 8-2 Static water contact angle showing a change in the wettability of

the surface after coating the surfaces with monolayers. .......................... 137

Figure 8-3 XPS survey spectra before (bottom) and after (top)

functionalisation of a ODPA SAM coated surface. ................................ 140

Figure 8-4 XPS survey spectra showing the change in the surface chemistry

after functionalising 16-PhDA SAMs with Ciprofloxacin®. Note the

presence of F 1s peak at 687.6 eV and increase in the intensity of N 1s

peak at 400.7 eV. ..................................................................................... 141

Figure 8-5 Deconvoluted a) C 1s, b) O 1s, c) N 1s and d) F 1s spectra obtained

using XPS for control, SAM coated and drug coated surfaces. .............. 142

Figure 8-6 Deconvoluted C 1s and O 1s spectra for Ciprofloxacin® powder

obtained using XPS. ................................................................................ 143

Figure 8-7 Static contact angle measurements using a contact angle

goniometer. a) Before cleaning; b) after cleaning; c) after 16-PhDA SAM

attachment and d) after immobilisation of 16-PhDA SAMs with

Ciprofloxacin®. ....................................................................................... 144

Figure 8-8 Relative intensity of C 1s, O 1s, N 1s and F 1s spectra obtained

using XPS for samples exposed to oxidative conditions for different time

intervals. .................................................................................................. 145

Figure 8-9 Static water contact angles obtained to determine the stability of

Ciprofloxacin® under oxidative conditions at different time intervals... 147

Figure 8-10 High resolution XPS spectra of C 1s, O 1s, N 1s and F 1s regions

showing their relative intensities after the immersion of Ciprofloxacin®

coated surfaces for different time intervals. ............................................ 149

Figure 8-11 Static water contact angle exhibiting a change in the surface

wettability after the immersion of Ciprofloxacin® coated surface in Tris-

HCl buffer solution for different time intervals. ..................................... 150

Figure 8-12 Percentage drug released at different time intervals determined

using UV-Visual Spectrophotometer. ..................................................... 152

List of Figures xx

Figure 8-13 Antibacterial susceptibility test against S. aureus (a) and E. coli

(b). Label description: A - control standard filter paper disc without drug,

B - standard filter paper disc with drug; C - Control Ti6Al4V metal disc

without drug; D - metal disc coated with drug. ....................................... 154

Figure 8-14 XPS spectra of F 1s region for control, drug coated and drug

coated discs placed on S. aureus and E. coli bacteria. ............................ 155

Figure 8-15 Antibacterial susceptibility test against a) S. aureus and b) E. coli.

Label description: C- control disc with no drug; CD – control with drug

on; TC: discs coated with Tris-HCl buffer; 1, 2, 4, 6 – immersion time

intervals (in weeks) of the samples in Tris-HCl buffer solution. ............ 156

Figure 8-16 Inhibited zone diameters for S. aureus and E. coli. ................... 157

List of Schemes xxi

List of Schemes

Scheme 2-1 Reaction scheme showing the conversion of acetic acid to acetic

anhydride using thionyl chloride ............................................................... 43

Scheme 2-2 Scheme of reactions on acid chloride esterification of flufenamic

acid. Label description: Flufenamic acid reacting with thionyl chloride (1);

reaction intermediate (2); reaction intermediate reacting with water (3);

flufenamic acid (4). ................................................................................... 50

Scheme 4-1 General scheme of reactions to coat 16-PhDA monolayers to

Ti6Al4V surface and to functionalise the monolayers with

Ciprofloxacin®. Label description: Hydroxylated titanium surface (1), 16-

Phosphanohexadecanoic acid (2), SAMs adsorbed on titanium surface (3),

reaction intermediate formed as a result of acid chloride esterification (4)

and Ciprofloxacin® immobilised to SAMs. .............................................. 70

Scheme 7-1 Scheme of reactions to coat 16-PhDA monolayers to a Ti6Al4V

surface and to functionalise the monolayers with Paracetamol. Label

description: hydroxylated Ti6Al4V surface (1); 16-PhDA SAMs (2); 16-

PhDA SAMs chemisorbed to Ti6Al4V surface (3); reaction intermediate

(4); Paracetamol immobilised to SAMs (5). ........................................... 124

List of Tables xxii

List of Tables

Table 1-1 Some of the materials used for the fabrication of biomedical

implants [1, 4, 7, 8]...................................................................................... 3

Table 2-1 Definitions of AM processes categories in accordance to the ASTM

International Committee F42 on Additive Manufacturing Technologies

(2010) [62]. ................................................................................................ 16

Table 2-2 Variable parameters in laser scan strategy. ..................................... 22

Table 4-1 List of materials/chemicals used. .................................................... 60

Table 4-2 Summarised SLM Process Parameters ............................................ 67

Table 5-1 Calculated particle size distributions obtained for virgin, recycled

and sieve retained powders using a Malvern Mastersizer (average of 5

measurements). .......................................................................................... 82

Table 5-2 Relative atomic percentages obtained for virgin and recycled

powders using XPS. The calibration error during peak integration is ±

10%. ........................................................................................................... 84

Table 5-3 Relative atomic percentage of elements detected using XPS for

SLM-AF, SLM-MP and FGD-MP samples. The calibration error during

peak integration is ± 10%. ......................................................................... 90

Table 5-4 Relative atomic percentage of elemental contributions obtained

using an XPS. The calibration error during peak integration is ± 10%. .. 108

Table 5-5 Relative atomic percentage of elements detected during XPS

characterisation of Ti6Al4V powders and side surface. The calibration

error during peak integration is ± 10%. ................................................... 109

Table 7-1 Relative atomic percentages of the elements detected in XPS. The

calibration error during peak integration in XPS is ± 10%. .................... 121

Table 7-2 Relative atomic percentages of elements detected by the XPS in

SAM coated and Paracetamol coated Ti6Al4V surfaces. The calibration

error during peak integration in XPS is ± 10%. ...................................... 125

Table 8-1 Relative atomic percentage of the elements detected by XPS after

different surface treatments. The calibration error during peak integration

in XPS is ± 10%. ..................................................................................... 135

List of Tables xxiii

Table 8-2 XPS determined relative atomic percentages of elements detected

for different oxidative exposure time intervals. The calibration error

during peak integration in XPS is ± 10%. ............................................... 146

Table 8-3 XPS determined relative atomic percentages of elements detected

for different immersion time interval in Tris-HCl buffer solution. The

calibration error due to XPS peak integration is ± 10%. ......................... 148

List of Symbols and Units xxiv

List of Symbols and Units

List of Symbols Used for Chemicals

Al Aluminium

Ba Barium

C Carbon

Cl Chlorine

Co Cobalt

Cr Chromium

F Fluorine

Mg Magnesium

N Nitrogen

O Oxygen

P Phosphorous

S Sulfur

Si Silica

SS Stainless Steel

Ti Titanium

V Vanadium

List of Units

°C Degree Celsius

λ Wavelength

µm Micrometre

µs Microseconds

amp Ampere

cm Centimetre

cm2

Square centimetre

eV Electron volts

g Gram

GPa GigaPascal

MHz Mega Hertz

mL Millilitre

mm Millimetre

mm/s Millimetre/second

nm Nanometre

mm2

Square millimetre

N Newton

Pa Pascal

ppm Parts per million

V Volt

W Watts

Special Characters

® Registered

G+ Gram Positive

G- Gram Negative

List of Acronyms xxv

List of Acronyms

16-PhDA 16-phosphanonexadecanoic acid

3D Three Dimensional

ACE Acid Chloride Esterification

AFM Atomic Force Microscope

AM Additive Manufacturing

ASTM American Society for Testing and Materials

ATCC American Type Cell Collection

ATR Attenuated Total Reflectance

CAD Computer Aided Design

CARTIS Centre for Applied Reconstructive Technologies in Surgery

CT Computed Tomography

CNC Computer Numeric Control

DCM Dichloromethane

DC Direct Current

DE Direct Esterification

DES Drug Eluting Stents

DHE Direct Heat Esterification

EDM Electrical Discharge Machining

EHT Extra-high Tension

ESCA Electron Spectroscopy for Chemical Analysis

FGD-MP Mechanically Polished Forged Part

FTIR Fourier Transform Infrared Spectroscopy

IGES Initial Graphics Exchange System

IR Infrared

MIC Minimum Inhibitory Concentration

List of Acronyms xxvi

MRI Magnetic Resonance Imaging

NIS National Inpatient Sample

NIST National Institute of Standards and Technology

NJR National Joint Registry

NSK No Skin Scan

ODPA Octadecylphosphonic acid

PMMA Polymethylmethaacrylate

SAMs Self-assembled Monolayers

SE Standard Error

SEM Scanning Electron Microscope

SK Skin Scan

SLM Selective Laser Melting

SLM-AF As-fabricated Selective Laser Melted Part

SLM-MP Mechanically Polished Selective Laser Melted Part

THF Tetrahydrofuran

Tris-HCl Tris-hydrochloride

UK United Kingdom

USA United States of America

UV Ultra-violet

UV-Vis Ultra-violet Visible Spectroscopy

XPS X-ray Photoelectron Spectroscopy

Chapter 1: Introduction 1

1. INTRODUCTION

1.1. Biomaterials

Biomaterials are natural or synthetic materials (other than drugs) that are used

to evaluate, treat, augment or replace any damaged tissues/organs/function of

the body [1]. The field of biomaterials is of major importance to mankind since

their use has played a significant role in improving life expectancy and quality

of life [2].

The use of biomaterials is not new; Egyptians and Romans used linen for

sutures, gold and iron for dental applications and wood for toe replacements

[3]. After the Second World War, nylon, teflon, silicone, stainless steel and

titanium were some of the other materials introduced for biomedical

applications. With recent advancements in human healthcare systems and with

the availability of improved diagnostic tools, biomedical implants have found

applications in almost all body functions (Figure 1-1). It is expected that

virtually every individual will have contact with biomaterials and/or

biomedical implants at some point of time during his or her life. This contact

may occur in several ways including (i) permanent implantation (such as heart

valves and total joint replacement); (ii) long-term applications (such as contact

lenses, dental prostheses and fracture fixation devices) and (iii) transient

applications (including needles for vaccination, wound healing

dressings/sutures and cardiac assist systems).

Biomedical implants range from simple wires and screws for fracture plate

fixation to joint prostheses for hips, knees, shoulders and so on. Depending on

the requirement and load conditions, implants are made of natural and/or

synthetic materials such as polymers, metals, ceramics or composites. For

example, metals are generally preferred for load bearing implants and internal

fixation devices whereas ceramics are preferred for skeletal and hard tissue

repair [4]. Polymers have been found the optimal material class when

requirements of little response to implantation, long-term stability in hostile

environment, low material stiffness leading to high material flexibility, and


good electrical insulation of metallic conductors have to be combined in a

single material [5].

Figure 1-1 Biomedical implants for human application. Reproduced with kind

permission from Bentham Open [3].

Biomedical implants are fabricated by employing one or more of the following

conventional manufacturing techniques including forming, and/or subtractive

manufacturing. In the forming technique, components are made in number of

ways such as casting a liquid material to a mould or a die, joining of materials

together by welding or knitting/braiding and by forging the material to obtain

the desired shape. Laser or water jet cutting, milling, electrical discharge

machining (EDM), computer numerical control (CNC) machining are some of

the subtractive manufacturing processes used to fabricate most of the currently

available implants for hip and knee replacements, vascular stents, dental and


maxillo-facial implants. These subtractive manufacturing methods are used on

their own or in combination with other forming or casting techniques.

1.2. Market for Biomedical Implants

Implants available for various body systems and the material with which they

are fabricated are tabulated in Table 1-1.

Table 1-1 Some of the materials used for the fabrication of biomedical implants [1, 4, 7, 8].

Application Materials used

Skeletal system

Joint Replacements (hip, knee

and shoulder)

Titanium, Ti-Al-V alloy, stainless

steel and ultrahigh-molecular-weight

polyethylene.

Finger joints Silicone.

Bone plate for fracture

fixation

Stainless steel and cobalt-chromium

alloy.

Bone cement Poly(methyl methacrylate).

Bony defect repair Hydroxyapatite.

Artificial tendon and ligament Teflon and Dacron.

Dental implant for tooth

fixation

Gold, Titanium, Ti-Al-V alloy,

stainless steel, polyethylene, alumina,

and calcium phosphate.

Maxillo-facial and cranial

prostheses

Titanium, Ti-Al-V alloy, stainless

steel, polydimethyl siloxane,

polyurethane, polyvinylchloride and

polyetherketoneketone.

Cardiovascular

system and other

vascular systems

Blood vessel prostheses Dacron, Teflon and polyurethane.

Heart valve Reprocessed tissue, stainless steel and

carbon.

Catheter Silicone rubber, Teflon and

polyurethane.

Stents Stainless steel, Nitinol®, magnesium,

Elgiloy®, tantalum, platinum-iridium

alloys, Poly-L-lactic acid, poly-D,L-

lactic acid, poly caprolactone and

polyglycolic acid.

Organs

Artificial heart Polyurethane.

Skin repair template Silicone-collagen composite.

Artificial kidney Cellulose and polyacrylonitrile.

Heart-lung machine Silicone rubber.

Senses

Cochlear replacement Platinum electrodes.

Intraocular lens Poly(methyl methacrylate), silicone

rubber and hydrogel.

Contact lens Silicone-acrylate and hydrogel.

Corneal bandage Collagen and hydrogel.

Other

Sutures Polylactic and polyglycolic acid

Gastrointestinal segments Nylon, polyvinylchloride and silicone

Breast enlargement Silicone.


According to the National Joint Registry (NJR), UK’s statistics for the year

2012 – 2013 (1st April – March 31

st) in England, Wales and Northern Ireland,

joint replacements including hip, knee, ankle, shoulder and elbow accounted

for having the highest ever annual number of 196,403 [6]. Nationwide

Inpatient Sample (NIS) in the United States of America (USA) shows that

primary hip replacements increased by 48% and first time knee implants grew

by 63% from 1997 to 2004. It is estimated that if these trends continue, the

current estimated 600,000 hip and knee replacements will increase to 3.4

million by 2030 in the USA [7,8].

Bibb et al. (2010) reported that in excess of 64,000 facial prostheses are made

annually worldwide. According to Datamonitor’s report (2006), the USA

accounts for 50% and Europe 30% of the total procedures worldwide [9]. The

2005 revenues for hip implants in the US were $2 billion and $1.4 billion in

Europe, while knee implant revenues comprised $2.4 billion in the US and

$774 million in Europe. According to Ratner et al. [1] the US market for

biomaterials accounted for $9 billion. These figures show that the use of

implants has become an integral part of our human health care system and a

big rise in the biomedical sector over the last few decades.

1.3. Biological Response to an Implant

Understanding the mechanism by which cells respond to an implant is of major

importance to achieve long-term success of the implant surgery [10]. Almost

all biomaterials implanted in a living tissue/system undergo tissue responses

within a few seconds after implantation [11]. On implantation, the implant is

surrounded by water molecules, which trigger the host proteins. The surface

property of an implant plays an important role at this stage on the extent and

specific interaction of the implant’s surface with this hydration layer. Based on

this interaction, the host proteins respond to the implant surface and form a

thin layer of extracellular matrix protein film [10].

The presence of proteins including albumin, fibrinogen, fibronectin,

vitronectin, in the protein film modulates the host inflammatory cell interaction

and adhesion [12]. Hence, the protein film is responsible for the control of

subsequent biological reactions (such as inflammatory and wound healing


responses) at the implant-host interface [4]. Figure 1-2 schematically

represents the sequential reactions that take place after the implantation of a

biomaterial into a living system [4].

Figure 1-2 Schematic of the sequential reactions that take place after the implantation of a

biomaterial. Figure reproduced with kind permission from Elsevier [4].

The adherence of different proteins, its concentration and conformations are

controlled by the surface properties of the implanted biomaterial. The

interaction of the adsorbed proteins with adhesion receptors such as (integrin)

present on inflammatory cell populations constitutes the major cellular

recognition system for implants [12]. The cell-protein interface formed after

implantation will initiate cell adhesion, migration and differentiation. Some of


the factors that govern cell adhesion and growth are extracellular matrix

proteins, cell-surface bound proteins and chemical characteristics and

topography of the implant surface [13].

Since the surface properties of an implant is critical in determining the long-

term success of the implant, there is a significant interest in optimising the

implant-host interface by altering the surface properties specific to the site of

implantation. For example, for cardio-vascular stent applications,

haemocompatibility of the stent material would be advantageous whereas for

the bone replacement applications, cytocompatibility of the implant would be

beneficial. Surface modification of biomaterials is discussed in section 2.4.

The final stage of the body responses to an implant can last up to several

decades. For example in an orthopaedic application, as a result of body

responses, eventually a functionally active mineralised bone is formed

surrounding the implant. Adverse reactions or post-implant complications such

as fibrous capsule formation, granulation tissue formation, implant failure due

to fracture and acute/chronic inflammation can happen at any time after

implantation.

1.4. Issues with Current Biomedical Implants

Biomedical implants have significantly improved the quality of life for

countless people; however, they still impose significant challenges on current

manufacturing processes due to their need to function within a relatively harsh

biological environment [2]. Customisation and fabrication of complex

geometries, and the reduction of post-implant complications including implant

migration, fracture and poor biocompatibility are major challenges to

overcome.

1.4.1. Customisation and fabrication of complex geometries

Most off-the-shelf implants are available in a selection of dimensions to meet

the patient requirement; however, they are not customised. Customisation is

key in conditions such as where the patient’s anatomy deviates from the

standard sizes or is affected by individual defects. The primary objective of

customisation is to fit the unique anatomy of a particular patient, especially


where the implant is not fixed. Personalised hearing aid shells for “in-the-ear”

and “in-the-canal” devices are one example of a customised implant that has

achieved superior fit compared to conventionally manufactured aids increasing

both comfort and device functionality [14]. Orthopaedic implants usually

perform better when they match exactly the anatomy of the patient, through the

distribution and normalisation of the stresses incurred in the remaining skeletal

system and reduction of migration and failure [15].

Migration of an implant is the movement of the implant from the actual or

surgically positioned area and is mechanically triggered rather than a

biological process [16]. Forces higher than expected may deteriorate fixation

and cause implant migration. Migration of an implant to a more harmful

anatomical space can sometimes be life threatening. Implant migration has

been reported for a number of devices including total hip and knee

replacement, dental, maxillo-facial and vascular implants [17–21]. Some of the

factors that determine the rate of migration other than mechanical forces are

implant design and fixation methods [16].

Cementing, screw retention of the implant, expanding the implant to conform

to the walls and press fitting of the implant to the diseased/damaged area are

some of the common fixation methods. However, all of these retention

methods have advantages and limitations over the other methods and selection

of an appropriate fixation method is key in long-term applications [22–25]. For

example in orthopaedic surgery, polymeric cements such as polymethylmetha-

acrylate (PMMA) is used to fix prosthesis due to its strong mechanical

fixation. However, when using this PMMA cement fixation for long term

application, loosening of implants due to a loss of mechanical fixation is

observed. In recent years, bioresorbable bone cements are used to provide both

mechanical fixation in the early stages leading to biological fixation [26].

Biological fixation is attained due to osseointegration after the resorption of

the bone cement. Press-fit implants are considered as an option in hip

replacements for patients with strong bone (since they are most likely to have a

high possibility for biological fixation) and cement-retention are considered

where there is a concern about bone quality. However, screw retention, press-


fit and cement restorations are practiced in dental applications depending on

the need and medical conditions [27].

Fixation of implants may sometimes be difficult depending on the site and

anatomy of the diseased/damaged part. In such cases, customisation may

render improved fixation with the patient’s anatomy and may normalise

stresses thus preventing implant loosening and migration. However, the ability

to achieve customised and complex geometries of biomedical implants using

subtractive or formative manufacturing methods (such as moulding, die casting

or subtractive processes) are limited and are often time consuming [27]. This is

because these methods have limitations in the geometries that they can

produce. Moulding and die casting methods require dies or moulds to fabricate

the part. Hence, extra time for the manufacturing process is required in

addition to fabricating the actual part. In addition, it is expensive to make a die

or mould for a one-off-design. With subtractive manufacturing techniques such

as computer numeric control (CNC) machining and laser cutting, parts with

complex internal and external geometries are sometimes difficult to achieve.

1.4.2. Post-implant complications

In 2012 alone, among the total of 86,488 hip replacement procedures in

England, Wales and Northern Ireland, over 10,000 were revision procedures

[6]. This revision procedure accounts to 12% of total hip replacement in 2012

whereas in 2011 it was 11%. There are many reasons for the failure of an

implant within the biological environment including manufacturing, chemical,

mechanical, tribological and surgical failures [3]. The patient’s health

condition and the physician’s experience could also be contributing factors. A

variety of biomaterials (polymers, metals, ceramics and composites),

fabrication techniques (such as compression moulding, die casting, bar stock

milling and laser cutting) and surface modification techniques (biocompatible

material coating, surface polishing and drug coating) have been employed to

improve the mechanical and biological properties of the implant; however,

there are still constraints in achieving this objective.


Even when using a biomaterial with suitable material and biological properties,

there are possibilities for the failure of an implant due to faulty mechanical

design or inappropriate application of the implant. Inadequate mechanical

properties (e.g. elastic modulus, yield strength, tensile strength) can result in

fracture leading to implant failure [17,28,29]. Elasticity modulus is the

resistance offered by an object/substance when deformed elastically. Yield

strength is the stress at which a material starts to deform plastically. Tensile

strength is the maximum stress which the material can withstand before its

failure. Fracture of an implant mainly occurs when the implant is not capable

of bearing the load exerted on it. For example, the major purpose of

orthopaedic implants is to restore the function of load bearing joints (such as

hip or knee joints) that are subjected to high levels of mechanical stresses,

wear and fatigue during the course of normal activity [4]. According to Wolff’s

law, the bone of a healthy human or an animal will adapt to the loads under

which it is placed. For example, when there is an increase in loading of a

particular bone, the bone will remodel itself over time to resist that load. The

internal structure of the trabeculae will undergo adaptive changes, leading to

secondary changes to the cortical portion of the bone. However, it should be

noted that if the loading on a particular bone decreases, then the bone will

become weak.

Although the mechanical properties of the bulk material, implant design and

manufacturing process are considered to be the major factors determining

implant fracture, corrosion of implants within the biological environment is

also a crucial factor [3]. Before the advent of 316 stainless steel, metallic

implants of various steel formulations failed dramatically due to tissue

reactions [30]. 316L SS have a better corrosion resistance and relatively low

carbon concentration in the alloy mixture compared to other steels. The

presence of molybdenum increases the corrosion resistance of the alloy.

Implants corrode within the biological environment due to electrochemical

attack by the electrolytes present in the hostile environment. Body pH can

affect the corrosion resistance of implants. Although the pH value of the

human body is normally maintained at pH 7, this might change from pH 3 – 9


depending on the imbalances in the biological system due to infections,

diseases and other factors. The aqueous medium in the human body consist of

anions including chloride, phosphate and bicarbonates, and cations including

sodium, potassium, calcium and magnesium [3]. Leaching of metal ions from

the implant surface due to corrosion will lead to erosion. The rate of corrosion

is accelerated by increased surface area and loss of protective oxide film, later

leading to fractures. Also, leaching of metal ions has been observed to induce

acute and chronic effects. For example, the release of nickel from stainless

steel was observed to affect the skin; cobalt was observed to cause anaemia

which inhibits the absorption of iron into the blood stream; chromium was

observed to cause ulcers and disturb the central nervous system; aluminium

was reported to cause epileptic effects and Alzheimer’s disease and leaching of

vanadium in its elemental state was observed to be toxic [31]. The tolerable

corrosion rate for metallic implants is 2.5 x 10-4

mm/year [3]. Hence, corrosion

products formed as a result of implant-host interaction have a significant effect

on the long term stability of the prosthesis and its cytocompatibility [32].

Biocompatibility, a requirement for all biomedical implants, is the property of

a material to be compatible with the tissues in the human body by not

producing a toxic, injurious or immunological response [4]. Chronic

inflammations, lesions and scarring are some of the adverse reactions that may

take place in the site if the implant is not biocompatible. To achieve better

biocompatibility, two routes are often employed: (i) using biocompatible

materials such as gold; (ii) fabrication of implants using off-the-shelf materials

with application of a suitable biocompatible layer such as bio-active glass,

calcium phosphate and protein coatings [15]. The corrosion resistance of

titanium is better than other metals such as 316L SS due to the presence of a

stable surface oxide layer (TiO2) of approximately 10 nm. However, the

thickness of this surface oxide layer can be enhanced by various surface

modification methods including chemical etching and anodisation [30].

Contamination is another factor contributing towards an implant failure.

Contamination of implants such as stents, dental, hip and knee implants by

harmful microbes has been widely reported [33–36]. Bacterial infections due to


implants are one of the main risk factors in orthopaedic surgery leading to

surgical removal of implants [17,37,38]. Contamination of an implant is

possible due to non-sterile manufacturing and packing, during surgical

placement and/or even due to the microbes present in the body after

implantation into the body [38]. By entering the host through contaminated

surfaces, bacteria multiply in the host environment and interfere with the host

defence system leading to host tissue damage and inflammation [38].

Staphylococci species, including Staphylococcus aureus and Staphylococcus

epidermidis are the more common cause of implant-associated infections [39].

Treatment measures available for bacterial infections include systemic,

antibiotic loaded cements and antimicrobial coatings [37]. However, the

release of drugs in a systemic therapy prevails as one of the major problems

when conventional coating techniques are used. Slower release of antibiotics

for a longer period of time is a potential cause of antibiotic resistance in the

human body [40]. For example, in drug eluting stents (DES), to reduce the

restenosis rate, the stents are coated with polymers containing a drug using

simple techniques including dip/spray/spin coating or solvent casting [41,42].

After implantation, the DES is expanded to conform to the vessel wall. During

this expansion, fractures, peeling of the polymeric coating and uneven release

of the drug material were reported as some of the major problems with surface

modified DES [43,44]. To address these issues, Mani et al. [45] adopted a

coating technique by which the drug was directly loaded on to the implant

surface. However, the amount of drug coated was very low (~ 5µg/cm2)

compared to the amount of drug coated on a commercially available drug-

eluting systems (100 µg/cm2). Polymer brushes were also used to coat implant

surfaces with drugs [46]. Since these coating methods are in nano-scale, they

add only a few nanometres to the implant surface and hence, peeling of drug

coating is less likely to occur [47]. Also, implants coated with conventional

coating techniques have drawbacks including surface heterogeneity in the type

and distribution of functional groups, hydrophilic and hydrophobic domains

and surface roughness [48].


1.5. Current Research

In order to address these prevailing challenges, a large amount of research is

being performed worldwide aiming to reduce implant-associated complications

and improve quality of life. Development of biocompatible and biodegradable

materials, design optimisation of implants, the use of novel manufacturing and

surface modification techniques are explored widely. With the properties of a

material being a key factor in determining the success of an implant, there has

been a considerable interest in material development. Currently the use of

biodegradable materials, such as polycaprolactone and polyglycolic acid, is

popular since their use reduces risks associated with implants being in the body

for the rest of the patient’s life. Implants fabricated using biodegradable

materials can disintegrate after a pre-determined time interval which reduces

the need for surgical removal of implants.

Optimisation of the implant designs to bear specific loads has been researched

widely in recent years [49,50]. Optimisation methods allow the prediction of

areas of fracture/failure of implants at specific loading conditions. Thus,

implants can be designed to withstand the required load conditions with less

material usage. Conventional manufacturing techniques used to fabricate

biomedical implants have design constraints. Selective laser melting (SLM), a

metal-based additive manufacturing (AM) technique, is of particular interest

due to its capability to fabricate functional components having mechanical

properties comparable to those of the bulk materials [51]. The ability of SLM

to build complex parts from three dimensional (3D) computer designs

(including well-defined porous structures) also offers the possibility of

customisation of biomedical implants. The potential for using SLM to fabricate

biomedical implants has been discussed in more detail in the literature review

section.

Although customisation and mechanical properties of implants are very

important for avoiding post-implant complications, surface properties

including surface texture, surface chemistry and the stability of the surface

oxide layer have also gained considerable attention [4,13]. In recent years,

surface modifications have been performed not just to improve the surface


finish and corrosion resistance, but also to make the implant biocompatible and

a drug carrier. Also for certain applications, surface modification is required to

make an implant’s surface specific-to the binding of extracellular matrix such

as fibrinogen and laminin in order to promote cell adhesion and tissue

regeneration [4]. There are numerous surface modification techniques

(including coating with biocompatible materials or drug-eluting polymer)

available to improve the corrosion-resistance, cytocompatibility, bone-tissue

integration, drug delivery and reduce post-implant complications; however,

most of them have limitations and do not have precise control over the surface

chemistry [52–54]. Thus a coating technique that offers precise control on the

location and orientation of chemical groups/biomolecules on the surface is

essential to cater for current needs.

Self-assembled monolayers (SAMs) are composed of amphiphilic molecules

with a head group which can attach to the surface to be modified and a tail

group containing the functionality which is desired to be displayed on the

surface. There are different types of SAMs including thiols, silanes and

phosphonic acids. SAMs can be modified with various functionalities

including drugs and proteins. Also, SAM coatings are usually in nano-scale

and can be used to specifically modify a material’s surface. The ability of

SAMs to precisely control surface chemistry with relatively simple and

inexpensive processing, has led to numerous suggested applications and

biomedical applications is one among them [45,54–59]. A detailed explanation

of SAMs, their types and formation are discussed in section 2.5.

1.6. Scope of the Research

The key to address the challenges with biomedical implants is to optimise the

implant-host interface. It is envisaged that customised implant designs, due to

their superior fit compared to traditional designs, can reduce post-implant

complications by preventing implant migration. In addition, by modifying the

surface chemistry of the implant with drug/protein molecules specific to the

target site, biochemical interactions at the implant-host interface can be

optimised and post-implant complications can be reduced. Previous research

has demonstrated the use of SLM for producing biomedical implants with


customised and complex external and internal structures [60,61]. In addition

SAMs, a form of nano-coating, have the ability to modify the surface

chemistry of an implant precisely and specifically. This thesis will evaluate the

potential to integrate SLM with SAMs-based surface modification to fabricate

customised biomedical implants with functional properties.

1.7. Thesis Structure

Chapter 2 will discuss the available literature and help the reader to understand

the background knowledge in additive manufacturing (AM), in particular the

selective laser melting (SLM) process, as well as surface modification using

self-assembled monolayers (SAMs). This chapter will critically analyse the

literature and sets the aim and objectives for this research. Chapter 3

showcases the novelty, aim and objectives of this research. Chapter 4 describes

the research methodology. It will list the materials, equipment, procedures and

characterisation methods used in this study. Chapter 5 will present the results

obtained for the fabrication and surface characterisation of parts produced

using SLM. In this chapter the surface chemistry is investigated in detail by

comparing the surface chemistry of a SLM fabricated part with a

conventionally forged part. Further, the impact of laser power on the surface

chemistry is also studied. Chapter 6 will describe the results for the surface

modification of SLM structures using phosphonic acid monolayers. The

stability of these monolayers under in vitro condition on both SLM as-

fabricated and mechanically polished surfaces will be discussed. Chapter 7 will

present the attachment of Paracetamol as a model drug to functionalise the

SAMs and Chapter 8 will discuss the results obtained for the immobilisation,

drug quantification, oxidative and in vitro stability and antibacterial

susceptibility of Ciprofloxacin® attached to monolayers. Chapter 9 concludes

the work with suggestions for future work.

Chapter 2: Literature Review 15

2. LITERATURE REVIEW

2.1. Introduction

The purpose of this chapter is to introduce the reader to additive manufacturing

(AM) methods (in particular selective laser melting) and surface modification

using self-assembled monolayers (SAMs). This chapter begins by describing

the standard definitions and manufacturing techniques belonging to AM. Later,

SLM, a metal-based AM technique is introduced and previous literature on the

SLM process, processing issues that exist and biomedical applications of SLM

are discussed. In the second part of the literature review, issues with some of

the current surface modification techniques are debated. An introduction to

SAMs, their types, assembly methods and the advantages of phosphonic acid

SAMs for biomedical applications are discussed in the later part. Finally from

the literature, the knowledge gap in both of these SLM based manufacturing

and SAMs based surface modification methods for biomedical applications is

identified.

2.2. Additive Manufacturing

Additive Manufacturing (AM) has developed from the early days of rapid

prototyping to enable the production of end-use parts. According to the

American Society for Testing and Materials (ASTM), AM is defined as [62]:

“The process of joining materials to make objects from 3D model data,

usually layer upon layer, as opposed to subtractive manufacturing

technologies”.

Although there are many terms used to describe AM, the most commonly used

terms are free-form manufacturing, rapid manufacturing, additive fabrication,

additive layered manufacturing and 3D printing.

AM was first developed for polymeric materials and now all types of materials

including metals, ceramics and composites can be processed using these

techniques to varying levels of success [63]. Currently there are many different

AM processes available, most working on a layer-by-layer basis. However,

they differ through the form of the starting material and the mechanism used to


consolidate it. Some of the consolidation methods used include the use of a

laser or electron beam to selectively fuse polymer or metal powdered material;

curing of liquid resin with ultraviolet (UV) light; extrusion of molten polymers

from traversing nozzles; jetting of droplets from an array of nozzles, similar to

inkjet printing and console`dation of sheets of materials using ultrasonic

vibrations. Table 2-1 lists and defines the categories of AM processes in

accordance with the ASTM International Committee F42 on Additive

Manufacturing Technologies published in 2010 [62].

Table 2-1 Definitions of AM processes categories in accordance to the ASTM International

Committee F42 on Additive Manufacturing Technologies (2010) [62].

Category ASTM Definition

Vat Photopolymerisation An additive manufacturing process in which liquid

photopolymer in a vat is selectively cured by light-activated

polymerization. Example: Stereolithography.

Material Jetting An additive manufacturing process in which droplets of build

material are selectively deposited. Example: PolyJet.

Binder Jetting An additive manufacturing process in which a liquid bonding

agent is selectively deposited to join powder materials.

Example: ProJet 4500.

Material Extrusion An additive manufacturing process in which material is

selectively dispensed through a nozzle or orifice. Example:

Fused deposition modelling.

Powder Bed Fusion An additive manufacturing process in which thermal energy

selectively fuses regions of a powder bed. Example: Selective

laser melting.

Sheet Lamination An additive manufacturing process in which sheets of

material are bonded to form an object. Example: Ultrasonic

consolidation.

Direct Energy Deposition An additive manufacturing process in which focused thermal

energy is used to fuse materials by melting as they are being

deposited. Example: 3D laser cladding.

AM has gained considerable interest in recent years due to its ability to build

parts of complex geometries from three-dimensional (3D) computer-aided

design (CAD) model data [64]. AM is capable of building parts of almost any

geometry in a single step regardless of the complexity of the part whereas

conventional methods often need further steps and time or it is impossible to

do so [65]. Furthermore, the use of AM speeds-up the whole development

process since there is no need for dies/moulds and toolings [66]. Hence, the

use of AM to fabricate customised one-off parts is faster and cheaper than


traditional methods. However, still there are issues in terms of the scalability of

the AM processes to make large volume of parts due to its repeatability and the

materials that are available. There are concerns over the mechanical properties

such as hardness, ultimate tensile strength, surface roughness and density of

the parts fabricated by AM. Also, the process is not currently cost-effective for

large-scale production. Some of the active industries for AM include

aerospace, automotive, medical, architectural, games, military, art, sport,

construction and education; however, there has been particular interest for AM

in aerospace and biomedical industries owing to the possibility for high

performance parts where cost is not a main objective [64,67].

2.3. Selective Laser Melting

Selective laser melting (SLM) is an AM technique capable of fabricating

metallic parts. Being an AM technique, increased geometrical freedom allows

the fabrication of a part as envisioned without the manufacturing constraints

prevailed with conventional techniques such as CNC machining, moulding and

casting. Stainless steel (SS) was the initially used material when the SLM

process was introduced in 1999. However after its introduction, a significant

number of metals and alloys were developed for processing. Among stainless

steel, 316L and 17-4PH are commonly prescribed for the use in a SLM process

[68,69]. Badrossamay et al. [70] and Hauser et al. [71] have studied the use of

tool steel to make highly dense parts. The use of aluminium alloys including

AlSi10Mg [72] and AlSi12 [73] cobalt-chrome (ASTM75) [74], titanium and

its alloys (Ti6Al4V, Ti24Nb4Zr8Sn and Ti6Al7Nb) [75], Inox 904L [76] and

Inconel (625 and 718) [77] for SLM process were also reported in literature.

Apart from these materials, magnesium [78] and nitinol (an alloy of nickel –

titanium) [79] are also being researched to be processed in a SLM machine.

2.3.1. Generation of a CAD file and supports for SLM

In SLM, parts are built directly from 3D model data. Normally, the CAD file

for the SLM process is designed using the commercially available design

software including ProEngineer, Solidworks and Catia. For medical

application, the CAD file of the implant to be fabricated is normally generated

either by designing the part using CAD software packages or by reconstructing


the structure of the damaged part from magnetic resonance imaging (MRI) or

computer tomography (CT) data. A set of CT/MRI data can be combined using

commercially available software such as Mimics (Materialise NV) to convert

the 2D DICOM files to a 3D surface tessellated (STL) file. During the STL file

conversion from other formats such as initial graphics exchange specification

(IGES), the geometry is converted to a triangular mesh, which represents the

surface of the part. In some cases, CAD files can also be reverse engineered by

laser scanning the damaged part.

SLM is capable of building net-shape parts with complex geometries;

however, there might be some areas of the part that overhang and may need a

support material during building. In such cases, supports should be generated

in a way that they are easy to remove without damaging the part’s quality

during post-processing. For example, AM 250 (Renishaw Plc.) machines use

the same material for supports as they do for fabricating the part (not only AM

250, all SLM systems use the same material for the build and its supports).

However, they use a lower laser power for the supports than they use to

fabricate the actual part so that the supports can be easily removed. Also it is

important to position supports in a way that they can be removed easily and

they cover all major overhangs. Supports are also used to stop warping of the

part. Once the part is designed, the CAD file is then exported in STL format.

Appropriate orientation is selected and supports are generated (where required)

to the STL design. The machine interface software is then used to slice the

STL file into several layers according to the predefined layer thickness along

the Z direction.

2.3.2. Equipment and Principle of SLM

A schematic of the SLM process is shown in Figure 2-1. A typical SLM

machine consists of a scanning laser (usually a fibre optic laser with

wavelength λ=1070 nm) a hopper attached to a wiper (recoater), an elevator

that lowers a build platform to adjust the layer thickness, and a lens that

focuses the laser to the build area. Before starting the build, the chamber is

made to be inert by pumping in an inert gas such as argon. The powder from

which the part is to be fabricated is spread over the build platform from the


hopper to a pre-defined layer thickness. Depending on the surface quality and

fabrication speed requirements, this layer thickness can be fixed to be between

20 µm and 100 µm.

Shortly after a layer of powder has been spread, the laser beam scans the

powder in the areas specified by the layer of the model file and fuses them.

Once the scan is complete, the build platform moves downwards by a pre-

defined layer thickness for a new layer of powder to be spread over the

previously scanned layer and this process continues until the part is completed.

Once fabrication is complete, the build platform is raised and the part is

removed from the substrate. The non-fused material on the build area can be

sieved and recycled [63]. The part is normally sonicated with deionised water

for 30 minutes to remove non-sintered particles.

Figure 2-1 Schematic of the SLM process. Reproduced with kind permission from Vision

Systems [80].

2.3.3. Factors governing the SLM process

The common factors that influence a part’s physical and mechanical properties

include material properties, powder bed property, laser energy input strategy,


part building strategy and atmosphere of the build chamber [81]. Most of these

process parameters are variable but some are not. They are discussed below.

Material Properties

The SLM process uses powdered metal/alloys to produce parts. Although a

range of metals can be processed using this technique, the material properties

including particle shape and size distribution, absorptivity, reflectivity, melting

and boiling point, conductivity and other thermal properties should be

considered before processing of a powder material since they have a significant

effect during the SLM process.

Particle shape and its size distribution have been shown to effect flowability,

powder bed density and fluidity of the material [72]. This may lead to

increased porosity as the larger particles melt or partially melt while the

smaller particles vaporise during the process [82]. Kempen et al. [72] reported

that varied powder morphology and its size distribution affected the SLM

powder bed density by 1%. Particle size distribution of the material directly

influences the powder bed density. In the SLM process, metal powder is spread

by a wiper. As the wiper moves across the bed, it may deliver certain pressure

on the powder bed increasing the powder bed density. Hence, the powder bed

density in a SLM process may lie between apparent density and tapped density

of the powder [81].

In the SLM process, a layer of powder should be evenly spread over the

platform to a predefined thickness for the laser to scan and fuse the material. If

the layer is uneven, it can have a significant effect on the melting process and

in turn affect the part’s quality. In order to achieve good flowability and

powder bed density, spherical particles are preferred. A wider particle size

range of 5 µm – 70 µm was observed to offer a higher powder bed density of

5.31 g/ml than the narrow size range of 20 µm – 50 µm (4.88 g/ml) [81]. This

may have been due to the fact that the smaller particles filled in the gaps

between the relatively large particle sizes. As the packing density was

increased, porosity in the fabricated part reduced from 0.11% to 0.05% [83].

However, it should be noted that in addition to the particle size, morphology,

distribution and powder bed density, the physical and chemical properties of


the material are also the major driving factors and are discussed in the later

section. The use of fine powders may create a dust cloud and may potentially

cause respiratory disorders. Hence, any operation with the fine powders should

be performed in a closed chamber or controlled environment.

Optical Scanning system

An optical system in a SLM machine contains a short pulse or continuous

wave of laser energy, a beam expander, a dual axis scanning mirror and an F-

Theta focal length lens [77]. Commercially available SLM machines have laser

powers normally ranging from 50 – 400 W and sometimes even higher in

certain machines. Since metals have high reflectivity (the ratio of energy of a

wave reflected from a surface to the energy possessed by the wave striking the

surface) at long wavelengths of incident laser and reduce the process

efficiency, lasers with short wavelengths (such as the fibre laser) are preferred

for the SLM process. Most commercially available SLM machines use fibre

laser with a wavelength (λ) 1070 nm – 1090 nm [84]. They also use fibre

modulated continuous wave laser operated in a pulsed laser fashion. This is

because a continuous laser will produce a continuous and a stable energy to

heat and melt the material homogenously. This can build up heat and increase

melt-pool width. In contrast, a pulsed laser produces burst(s) of energy to melt

the target material and thus the laser energy is delivered in a way such that heat

build-up and melt-pool width can be minimised [77]. The use of an F-Theta

lens ensures that the delivered laser energy is constant regardless of position in

the powder bed.

Previous studies on the SLM process have shown that the surface roughness,

porosity, dimensional accuracy and microstructure of a part greatly depends on

the laser processing parameters including laser power, scan speed, scan

strategy, beam width or spot size and scan overlap [51]. Laser energy density

[the ratio of incident laser power (W) to the scanning speed (mm/s) and the

laser spot diameter (mm)] is a critical parameter in the SLM process. The

incident laser power and scan speed are controlled by the machine controller

and the beam spot size is controlled manually by adjusting the lens.


Laser scan strategy is one of the important factors that determine the part’s

quality. A list of other variable parameters used in the scan strategy is

tabulated in Table 2-2. Figure 2-2 is a schematic representation of some of the

variable parameters used in the laser scanning. Energy density of the laser is

mainly affected by the scan speed in addition to the laser power. Scan speed of

the laser beam can be calculated as the ratio of point distance (µm) to the

exposure time (µs). By varying the range laser power and scan speed same

energy density can be altered. However, depending on the powder properties,

the effect of energy density will be variable.

Table 2-2 Variable parameters in laser scan strategy.

Parameter Description

Exposure time The time of laser exposure to each point

Hatch type Scan strategy such as scan direction and scan type.

Hatch distance The distance between two adjacent lines of the scan

Point distance Each hatch line is divided into series of points. Point distance

is the distance between two points of a hatch line.

Scan speed The speed of laser beam movement while scanning.

Figure 2-2 Schematic representation of laser scanning related input process parameters (a);

SLM fabricated Ti6Al4V part showing actual laser scan track (b) and further magnified area

depicting point distance of the laser scan(c).

The aim of optimising energy intensity is to make sure that the heat absorbed

by the powder as a result of laser – powder interaction is enough to melt and

produce fully dense part without overheating. Overheating of the powered

material may lead to the formation of heat affect zones with mixed phases and

may cause vaporisation and/or segregation of its alloying elements [51]. By

reducing the laser power and increasing the hatch distance, the energy density

can be reduced. Similarly when the scan speed is increased, the laser beam has


a short interaction time with the powder material and will lower the energy

density.

Several laser scan strategies are available to consolidate the powders including

uni-directional, bi-directional, alternating bi-directional, multi-directional and

meander. Figure 2-3 shows a schematic representation of some of these scan

strategies. The impact of scan strategies on the quality of the SLM parts has

been discussed in the later section. Hence by altering the variable parameters

of the optical system, highly dense parts with a better surface quality can be

produced.

Figure 2-3 Schematic of some of the scan strategies typically used in a SLM process.

Unidirectional (a); bi-directional (b) and alternating bi-directional (c).

Build atmosphere

Build atmosphere has a major part in determining the quality of the parts built

in a SLM machine. In general, the build chamber atmosphere contains

nitrogen, oxygen, carbon dioxide and inert gases. In an SLM process, the metal

powders are melted and fused together at a very high temperature. During this

process, the molten metals will react with the gases present in its atmosphere.

If the SLM process is performed at atmospheric conditions, there is a high

possibility for the metals to form its corresponding oxides, carbides and

nitrides by reacting with the available oxygen, nitrogen and carbon in the

atmosphere. For example, if SLM of titanium or magnesium is performed at

atmospheric condition (in the presence of oxygen), since they are highly

reactive, the metal will oxidise and burn/vaporise even before the melting point

is reached. These reactions can have a significant impact on the next build

layer causing porosity, delamination and reduction of the molten material’s

wettability which affects the fusion of successive layers, and possibly affects


the microstructure [84]. Also processing some metallic materials such as

aluminium in atmospheric conditions can be problematic. Hence to avoid such

adverse chemical reactions affecting the part’s quality, the SLM process

chamber is filled with an inert gas such as argon (99.99%). To aid in laser

melting and remove moisture content from the powder material, the

substrate/build platform is preheated to a certain temperature well below the

melting point of the material. Marcu et al. studied the flowability of non-heat

treated and heat treated (500 °C) Ti6Al7Nb particles (D90 = 56 µm) using

Carney funnel flowmeter (5 mm diameter orifice). Their study accounted for

an increased flowability (s) of 7 for the heat-treated Ti6Al7Nb whereas, no-

flow of powders were observed for non-heat treated powders [85]. Thus pre-

heating of powders is useful to improve the flowability. However, as

mentioned in the material properties section, irregular particle shape and size

can affect the packing density to 1%.

2.3.4. Heat Transfer in a SLM Process

Laser-material interaction in a SLM process is directly influenced by the

ability of a material to absorb the incident radiation. However, during the SLM

process, heat or energy input depends on numerous factors including material

type, reflectivity, surface properties of the material (such as roughness and

chemical composition), laser wavelength, direction of the incident radiation

and temperature of the material.

Laser processing is usually a fast heating process. When the laser reaches the

surface of a powder bed, the radiation is absorbed, reflected and transmitted

through the bed. According to Fourier’s theory of heat transfer, heat in a

material flows via conduction due to the thermal gradients. In the SLM

process, heat generated by the laser is absorbed by the metal powder and is

then transferred to the particles in contact through conduction and convection.

Absorption of radiation in metals is influenced by various factors [84]. When

the incident radiation is perpendicular to the metal surface, the absorptivity

was observed to be low. Increased oxide layer film thickness of the metal,

surface roughness and temperature were observed to increase the absorptivity

of the corresponding metallic material. Some of the other factors affecting the


absorptivity of the material include material composition and surface

contamination. Although most of the energy is conducted through the particles,

some portion of the absorbed and reflected or transmitted energy is lost.

During the laser scanning process, the energy is mainly lost due to reflectance

of the beam and energy transfer to the powder bed. Whereas while melting, the

heat energy is lost due to radiation, thermal diffusion and vaporisation [84].

When heat is conducted to a particle, it reaches to the core depending on its

thermal conductivity. [81]. Thermal conductivity is a measure of rate of heat

transfer through the material. When the heat conducted through is sufficient, it

liquefies the metal particles and results in the formation of a melt pool [86].

Although the metallic powder materials have high conductivity, the powder’s

thermal conductivity depends on the metal’s intrinsic properties and the

number of contacts made between the particles. For example, the rate of heat

transfer between the powders can be improved by increasing the powder bed

density which makes more contacts between them. The bulk density of the

metal powder itself can be a factor affecting the thermal conductivity [87].

2.3.5. Melt Pool Dynamics

A melt pool is formed when the absorbed energy is sufficient to change the

solid phase of the metallic particles to liquid (Figure 2.4). As the laser scans

the powder particles, the energy is absorbed and it is conducted to the

surrounding particles in contact to melt them. After the laser moves from the

scanned area, the melt pool is formed as a result of heat transfer between the

particles. The shape of the melt pool in a SLM process tends to be flat [77].

However, this greatly depends on the wetting angle. When a high laser power

is used, wettability is increased and a flat melt pool is obtained [77].

Wettability (the spreading behaviour of a liquid on a solid) is an important

factor that determines the capillary stability of the melt pool. Liquid metals

have poor wetting properties leading to capillary instability i.e. splitting of the

molten metal into small entities in order to reduce surface tension variations

[84]. This balling phenomenon is shown in Figure 2-4b. Balling is also referred

as sphereodisation of the liquid melt pool [82]. Balling was observed to

significantly cause weak inter-layer adhesion leading to the formation of non-


uniform layers. Further to this, balling was also observed to affect the part’s

quality by inducing porosity and increasing the surface roughness [77].

However, balling can be controlled by optimising the process parameters such

as laser power, scan speed and hatch spacing [84].

Figure 2-4 Schematic representation of melt pool (a) and splitting of melt pool resulting in

balling phenomena (b).

2.3.6. Microstructural Development and Thermal Effects

As the laser scan is complete, the melt pool stabilises and solidifies with the

release of thermal energy. In the SLM process, both the melting and cooling of

the metal/alloy occurs rapidly [51]. Due to this, a non-homogenous nucleation

is observed. Temperature gradients and heat transfer conditions determine the

cooling rate, grain growth and the formation of microstructure [81]. The

microstructure of the part fabricated by SLM determines the mechanical

properties of the part such as hardness, elongation and fatigue behaviour.

Altering the SLM process parameters can produce varied grain size, grain

growth and distributions. For example, laser scan vector length was observed

to influence microstructural features including the width of the prior β grains

and the thickness of the α’ laths. On comparing the microstructural features of

meander and checkerboard laser scan strategies, the width of the prior β grain

for Meander scan was 103 ± 32 µm whereas for the checkerboard strategy it

was 383 ± 32 µm. The α lath thickness for meander scan was observed to be

0.5 ± 0.2 µm whereas for the checkerboard strategy, this was 0.9 ± 0.5 µm.

[88].


During SLM, which is a rapid melting and cooling process compared to

conventional forming processes, the metal stays in the liquid state for a

fraction of second before solidifying. This short time and thermal cycle rate

induces residual stresses in the fabricated part leading to dimensional

inaccuracies, delamination of layers and solidification cracks [64]. Shiomi et

al. reported residual stresses for parts built using chrome-molybdenum-steel

alloy (JIS SCM 440) ranging from 300 – 400 MPa for laser scan speeds of 4

mm/s, 6 mm/s, 8 mm/s and 10 mm/s [89]. It was also reported in this study

that, by heat treatment for an hour, 70% of the residual stress were relieved.

Dimensional inaccuracies can be due to shrinkage of the part on solidification.

The rate of cooling of the liquefied metal during the SLM process will

contribute to the grain size distribution and microstructural pattern.

Solidification cracks originate when the solidifying metal cannot accommodate

the thermal shrinkage and this greatly depends on the material properties of the

corresponding metal [84]. Solidification cracks generally originate between

grain boundaries or at the interphase between different metals. The level of

shrinkage can be controlled by optimising the process, such as preheating the

powder bed and reducing the laser beam spot size. Preheating the powder bed

is suggested to reduce solidification cracks and layer delamination. Residual

stresses generated due to high thermal gradients can be relieved by post

processing of the part [84].

2.3.7. Existing research on SLM

The SLM process developed at the Fraunhofer Institute (Germany) is believed

to have been first commercialised by Fockele and Schwarze (Germany) in

1999. However, it is still considered a novel/innovative manufacturing

technology and the process is constantly developing owing to its ability to

make end-use parts and growing interest in diverse applications. Bridging

SLM from research to mainstream manufacturing is not straightforward, as

demanding industry standards and compliance need to be fulfilled. Material

development, optimisation of processing methods to achieve better surface and

mechanical properties, process economics, developing design rules for the

process and exploring the use of SLM in various fields are some of the areas

that researchers have concentrated on over the past decade.


As mentioned earlier in 2.3.3, the quality of a fabricated part greatly depends

on the factors that govern the SLM process. Those governing factors can either

be material based parameters (shape and size distribution of the particle;

surface morphology, density of the particle and material composition) and/or

equipment related parameters (such as laser power, scanning speed, scan

strategy, beam/spot size, hatch distance, layer thickness, powder bed

temperature and the atmosphere within the equipment) [76]. Manufacturing

strategies including build orientation and scan numbers are also some

important factors that can affect the quality of the part [74].

With SLM being a powder bed process, the material properties of the powders

will play a significant role in determining the quality of the part. The

manufacturing process of the powder, particle shape and size distribution,

optical and heat transfer properties, chemical composition of the powder and

thickness of the deposited layer for each fabrication cycle have all been

observed to be important variables for SLM relating to the feed powder

material [76].

Li et al. [68] studied the use of water atomised and gas atomised 316L

stainless steel powders and reported that gas atomised powder rendered a

denser structure when compared to the water atomised powder. This was due

to the gas atomised powder having lower oxygen content than water atomised

powder. The presence of high oxygen content can induce porosity,

delamination and also affects the wettability of the molten metal. Since gas

atomised powders have low oxygen content, they are preferred. The study also

revealed that the gas atomised powder yielded a higher packing density than

the water atomised powder since most of the gas atomised powder particles

were spherical when compared to water atomised powder. However, it should

be noted that internal pores may be present in gas atomised powders. The

presence of these internal pores will lead to reduced apparent density and may

cause gas bubbling when the laser beam scans the material [90]. Hence, this

may also induce pores in the part fabricated by SLM. Presence of pores can

affect the density of the fabricated part and are not favourable in applications

where fabricating a dense part is the requirement.


Rombouts et al. [91] explored the effect of elements including iron, carbon,

oxygen, copper, silicon and titanium on the quality of SLM fabricated parts.

Their study revealed that the presence of oxygen in the build chamber

increased the melt volume due to exothermic oxidation of iron and negatively

affected the part’s quality. Entrapment of carbon monoxide and carbon dioxide

due to the oxidation of carbon was observed to induce porosity. Silicon and

titanium were also observed to induce porosity due to their ability to form

corresponding oxides and carbides. The energy absorption coefficient of

oxides is higher than the metal and hence the oxides absorb more energy and

increase the temperature of the metal, resulting in overheating. As the material

is overheated, the rate of oxidation increases and affects the surface wettability.

Since parts are produced additively in a SLM process, wettability of the

previously scanned layer will have a significant impact on the density of a part

[92].

The density of a component determines its mechanical properties. Similar to

other manufacturing processes, the major goal of the SLM process is also

usually to attain 100% dense parts. However, due to the lack of mechanical

pressure (as found in the moulding process), this goal is difficult to achieve

[51]. SLM is mainly characterised by temperature effects, gravity and capillary

forces while processing. Entrapment of air bubbles due to low solubility of

dissolved elements in the melt pool, oxidation of carbon present in the

material, poor packing density of the successive layers and poor wettability of

the molten material induces porosities that affect the part’s density [91].

Studies conducted by Yadroitsev et al. [74] on 316L stainless steel (L in 316L

represents low carbon in the alloy composition, i.e. 0.03% maximum) revealed

that the production of highly dense parts is greatly dependent on the processing

parameters such as laser power, scan speed and layer thickness. A balling

phenomenon (splitting of the melt pool into small entities/droplets) was

observed to significantly affect the densification behaviour in the SLM process

[68]. Balling due to poor wetting of the melt pool was observed when low laser

energy was used to melt a thick layer. As a result, agglomerates with pores

between the metallic balls were observed as shown in Figure 2-5.


Pores are generated due to various other factors in addition to poor wettability.

In SLM, high roughness peaks and valleys are formed after scanning every

layer. When a fresh layer of powder is laid on this pre-melted and solidified

layer, the peaks and valleys may prevent a homogenous distribution of the

powder particles. As a result, gases are entrapped into the layer. When

scanning such layers, the laser energy may not be sufficient to melt the layer

completely due to the increased depth at certain locations caused by these

peaks and valleys. If the powders are not evenly spread, entrapment of gases

from the build atmosphere is possible. On laser scanning, the entrapped gas is

superheated and expands rapidly removing the molten metal above it and thus

inducing pores of sizes typically ranging from 2 – 50 µm [51]. Also when the

hatch spacing is wide, pores were observed [84].

Figure 2-5 Micrograph showing a polished 316L SS surface depicting a balling induced pore.

Reproduced with kind permission from Elsevier [68].

Morgan et al. [92] reported that the density of cubic parts fabricated using

316L SS in SLM increased with decreasing scan speed, when considering scan

speeds between 100 - 200 mm/s. This is because, at a high scan speed, the

energy per unit length will be less. Hence, less material will be consolidated

and thereby porosity will be increased. Kruth et al. [51] studied various scan

strategies (uni-directional, bi-directional and multi-directional) and reported

that multi-directional scanning provided improved relative density. Also, the

selective re-melting of each layer used in this study was observed to improve

the part’s density up to 98 – 99% for Ti6Al4V.


Li et al. [68] experimentally showed that at a high laser power, low scan speed

with a narrow hatch spacing and thin layer thickness, highly dense parts can be

obtained. Apart from laser power and scan speed, the quality of a part is also a

function of beam size/width, pulse shape hatch distance hatch spacing and

layer thickness. However, all of these process parameters are interrelated

[76,77]. Yadroitsev et al. [74] studied the effects of processing parameters

including laser power, scanning speed and powder layer thickness on the

formation of single tracks. Their results concluded that, by choosing an

optimal process window (layer thickness – 25 µm, laser power – 50W and

scanning speed ranging from 80 – 200mm/s for 316L SS) and appropriate

strategy of SLM, complex parts with mechanical properties comparable to

wrought material can be manufactured. Optimisation of SLM process

parameters has shown the potential to produce parts with mechanical

properties comparable to their corresponding bulk material [93,94].

Researchers have optimised the SLM process parameters and obtained a

relative density of up to 99% for AlSi10Mg and 316L SS [51,72].

Vandenbroucke et al. [95] reported that their optimised SLM parameters

yielded part densities up to 99.98% for Ti6Al4V. Other than process

optimisation, some post-processing (such as annealing, selective re-melting of

the layers) is performed to improve the mechanical properties and surface

quality of the SLM fabricated parts [93].

The mechanical properties of a SLM fabricated part not only depends on the

material properties and density, but also the microstructure patterns and defects

due to process parameters. Microstructures of the SLM fabricated parts were

noted to differ in different directions due to the layer manufacturing. Since in

SLM the metal powders are melted and cooled rapidly, the temperature

gradient and heat transfer conditions determine the cooling rate, grain growth

and the evolution of microstructures [81]. Rombouts et al. reported fine

ferritic grains for Fe powder processed using SLM [91]. Simoneli et al.

reported fine acicular α’ grains throughout Ti6Al4V sample and prior β grain

boundaries [88]. Li et al. [68] obtained a dense microstructure for a hatch

spacing of 150 µm. More details on the microstructural patterns of SLM

fabricated parts are recorded in the literature [51,68,91,96].


Defects such as porosity, delamination of layers and solidification cracks

within a part can significantly deteriorate mechanical properties leading to the

mechanical failure of the component [68]. Craeghs et al. [97] and Lott et al.

[98] developed a feedback control system to monitor the effect of these

parameters on the melt pool dynamics. Mathematical models were also

developed to estimate the effect of process parameters on the local temperature

distribution in the laser-material interaction zone to produce defect-free parts

by SLM [99].

Although SLM has several advantages in terms of what it can do and what it

can offer, poor surface quality of the SLM fabricated parts is of major concern

[100]. Some of the major causes of surface roughness are the balling effect,

partially melted and entrained powder particles that adhere to the outer edge of

the solidified melt pool and the stepped profile observed due to layer-wise

fabrication. The other contributing factor to the increased surface roughness is

the surface temperature difference between the laser beam and the solidifying

zone caused by the motion of the laser beam [86]. Kruth et al. [51] suggested

selective re-melting of the top layer (Z axis) for a better surface finish. By

surface re-melting, the average surface roughness (Ra) of 12 µm for non-laser

re-melted surface reduced to 1.5 µm. However, it is still a challenge to obtain

an even roughness profile all over the part. Post-processing methods such as

mechanical (abrasive sand blasting and machining), chemical/electro-chemical

(acid etching, oxidation and electro-polishing) and thermal processes

(annealing, plasma spray) are performed to smoothen the SLM surface.

Although an additional process is required to smoothen the surface profile,

SLM offers the advantage of fabricating customised and complex designs in a

single step that are almost difficult to fabricate with the subtractive or

formative technique.

Although the advancement of SLM as a robust manufacturing process is held

back by limitations including the availability of materials, surface finish,

mechanical strength and repeatability, it has still established profound

applications in diverse fields including automotive (customised structures),

aerospace (fuel nozzles for jet engines made by General Electric company) and


biomedical (surgical guides, maxillofacial and porous implants for orthopaedic

applications) where cost is not the prime concern [27,63,101,102].

2.3.8. Fabrication of medical devices using SLM

As discussed in section 1, customised manufacturing together with mass

production is required to meet the growing demand of joint replacements

[103]. The availability of customisation has been possible with advances in

manufacturing technology, enabling low volume products to be achieved

efficiently and AM is envisaged to be the enabler for many types of

customisation [104]. The need for customisation of biomedical implants makes

AM techniques such as SLM attractive for fabricating implants.

SLM not only has the ability to produce complex parts with desirable shape

and structure but also has the advantage of establishing a closed process chain

from scanning a damaged part of the body to design and manufacture. For

example, bone replacement materials can be fabricated by scanning the bone

defect using an MRI or CT, designing the implant structure, then directly

manufacturing the individual implant. This closed process chain will have the

potential to offer custom-fitting implants to the damaged part’s anatomy [60].

Also the freedom of design offered by this technology enables the fabrication

of complex geometries of the implants, in terms of both external and internal

morphology. The challenges including the necessities to build complex parts

with thin walled sections, customised implant design with micro-porous

structures and the need to reduce the market time can be well addressed by

SLM. SLM fabricated Ti6Al4V parts have also been reported to have good

mechanical properties matching ASTM F136 standards concerning in vitro

testing [60]. Inter-connected porosity offered by the SLM process is of

significant interest for fabricating bone replacement implants [75]. With its

advantages over conventional techniques that impose constraints on what can

be moulded or machined, SLM will not only improve the product quality but

could be a potential solution for some of the current biomedical issues

including the fabrication of customised designs and complex geometries.


Development of porous metallic surfaces for load bearing orthopaedic

applications is being researched since they promote bone ingrowth and

osseointegration [105]. Osseointegration refers to a direct structural and

functional connection between ordered living bone and the surface of a load-

carrying implant. Porous structures are required to lower the elastic moduli of

the component similar to that of bone. This is because metals exhibiting higher

elastic moduli than that of bone have been identified as a major reason for the

loosening of an implant [106]. For example the Young’s modulus for bone is

20 GPa whereas for titanium it is 110GPa. Therefore several techniques are

employed to fabricate porous structures and SLM is one of them.

To achieve improved cell adhesion, osseointegration and well-defined

interconnectivity, pore size and the amount of pores on the implant surfaces are

important. For example, pre-defined pores of a specific size and configuration

that can be formed using SLM allow the growth of bone tissues through the

implant in order to achieve a strong connection. For bone implants, a

customised design with porous structure will increase the contact of the

implant surface to the bone and increase the ability of bone regeneration [60].

The typical surface morphology of an implant fabricated using Ti6Al4V in an

AM 250 (a SLM machine manufactured by Renishaw Plc.) is shown in Figure

2-6. As can be observed from the figure, the as-fabricated surface is porous due

to partial melting of particles. Fabrication of porous scaffolds for bone tissue

regeneration using SLM has been reported extensively in the literature [107–

109]. Patanayak et al. [106] fabricated a titanium metal analogue to human

cancellous bone using SLM. Traini et al. [110] fabricated a porous dental

implant using Ti6Al4V by SLM and concluded that SLM is an efficient

technique to fabricate porous dental implants that adapt better to the properties

of a bone. Porosity is a measure of the void spaces in a material. Zieliński et al.

[111] have fabricated porous structures using Ti alloy by SLM and deposited

hydroxyapatite for improved osseointegration and lifetime of load-bearing

implants. Although the SLM-produced surfaces will be highly advantageous in

this context, care should be taken that the partially melted particles do not

detach from the surface since this can potentially lead to acute and/or chronic

effects depending on the biological response and site of implantation. For


example, in bronchial stents, peeling of these partially sintered particles may

block the alveoli in the lungs. Similarly, for cardiovascular stents, the presence

of rough surfaces can affect the flow of blood and peeling of these particles

may block the artery and affect the blood flow.

Figure 2-6 Typical surface morphology of a part fabricated using a Renishaw AM 250

machine. Note that the SEM micrograph is tilted to approximately 45° to the right to its build

direction.

Wehmoller et al. [103] showed the ability of SLM to fabricate a lower jaw

model (including teeth, the mandibular joint and the canal of the mandibular

nerve), a spinal column model and a tubular bone femur. It was reported in

their study that, to fabricate these implants in stainless steel, the production of

the lower jaw required 24 hours, the spinal column required 25 hours and the

bone femur took 26 hours. Although the time required to fabricate these parts

in SLM may seem longer, due to the complexity of the part, it may be difficult

to cast these models or fabricate using the subtractive/forming methods.

The Centre for Applied Reconstructive Technologies in Surgery, CARTIS UK,

demonstrated the feasibility of the design and fabrication of a titanium

cranioplasty plate and a successful implant to restore the orbital floor and rim

produced in titanium by SLM. The implant was produced by the Centre for

Rapid Prototyping and Manufacturing at the Central University of Technology,

Bloemfontein, South Africa. Figures 2-7 and 2-8 show the demonstration

cranioplasty implant and the implanted maxillofacial implant that were

fabricated using the SLM process. Drstvenstek et al. [112] fabricated a

bespoke angular implant using Ti6Al4V in a SLM machine to clinically treat a


patient with a congenital asymmetry of a facial bone. Their studies have also

reported a clinical use of a mandibular implant fabricated by this technique.

Figure 2-7 Customised cranial implant fabricated using the SLM process. SLM fabricated

implant (a); Implant with support structures (b); Implant removed from support structure

showing inner geometry (c) and custom-fitting of the implant to the skull (d). [Implant design

and 3D model courtesy of CARTIS].

Figure 2-8 Customised maxillo facial implant fabricated using SLM. CAD design of a

customised maxillo-facial implant (a); Implants fabricated in a SLM machine (b); Implant with

its support structure (c); Custom fitting of implant to a model of the damaged part (d) [Implant

design and 3D model courtesy of CARTIS].


Tolochko et al. [113] showed the possibility to fabricate dental root implants

with the required structure, geometry and strength by SLM. A successful

clinical use of a custom-made root analogue for a dental implant fabricated by

this process has also been reported in literature. Almeida et al. [114] evaluated

the reliability and failure modes of Ti6Al4V implant structure fabricated by

two different technique, SLM and alumina blasted/acid etched. Their

mechanical tests observed no significant differences in terms of reliability and

fracture modes between these two approaches.

In summary, SLM has been shown to be able to fabricate 3D porous structures

with customised, complex internal and external geometries which is difficult to

achieve by conventional manufacturing [106]. Currently, LayerWise,

Renishaw Plc and 3T RPD are some of the companies actively involved in the

production of biomedical implants for cranial, maxillo-facial, orthopaedic and

dental applications.

2.4. Surface Modification

Material selection is crucial for any application since the performance of a

manufactured part will reflect the properties of the selected material. Finding a

material with optimum bulk as well as the desired surface properties for any

given application is rare and biomaterials are no exception [30]. Bulk alteration

of a material is not generally preferred since it is expensive, time consuming

and will impose limitations for the use of the material. Surface modification of

a solid material at either atomic or molecular level is therefore performed to

achieve the required surface properties without altering the key bulk properties

of the material. Surface modification is required to reduce biological rejection

and these can be attained by surface modifying an implant’s surface with

suitable body proteins as peptides.

Surface modification has wide applications including adhesive bonding

enhancement, modifying wetting and non-wetting characteristics, creating

micro-porous structures, altering the grain size and distribution, generating

uniform three-dimensional structures on the surface and optimisation of

surface chemistries of an engineered part. Although surface modification is


performed in a number of engineering disciplines, its application in the

biomedical field is considered to be unique since it improves the quality of life

of the patient. Several types of surface modification of biomedical implants

such as morphological, physicochemical modification and biological

modification using proteins and cells have been reviewed and reported by

researchers [4,30,115–118].

2.4.1. Issues with current surface modification techniques

The ultimate aim of modern surface modification techniques is to make the

implant surface biocompatible and to direct the biological healing response

[115]. To address this goal, morphological, physico-chemical and biological

modifications are performed. However, they all have limitations. In most

cases, a single surface modification technique is not enough to achieve the

desired surface property and hence two or more surface modifications are

performed to achieve better biological properties. For example, Kim et al.

[119] reported good adherence and spreading of human osteoblast cells for

sand blasted and acid etched surface. Though the use of two techniques

provided improved results in terms of cellular attachments, the use of two or

more techniques to modify the surface can be time consuming and expensive.

Implant coating using polymers impregnated with therapeutic agents has

limitations that include fissures, cracks and waviness (undulating pattern) of

the coating, inflammatory and hypersensitivity reactions as shown in Figure 2-

9 [52]. The cracking and peeling of these polymeric coatings on the stents may

be due to the expansion of the material beyond its elastic limit.

Most of the polymer and protein coatings are formed from large molecules

with several reactive sites and hence it is difficult to introduce and control a

specific chemical group. Chemical treatments with an acid or a base can offer

varied surface finish (with micro pores and groves) and the introduction of a

few acidic, basic and hydroxyl groups; however it is limited [120]. Heat

treatment will have significant impact on the physical properties of a part but is

limited to chemical modifications in conditions where the surface has to be

functionalised.


Figure 2-9 Micrographs of BiodivYsio (a, b) Taxus (c, d) and Cypher (e, f) stents

commercially available for cardiovascular stenting. The stents exhibit cracking and peeling of

the polymeric coatings on their expansion. Reproduced with kind permission from HMP

Communications [44].

2.5. Surface Modification Using Self-Assembled

Monolayers

Self-assembly is a process in which a disordered system of pre-existing

component forms an organised structure or pattern due to specific local

interactions among the components. Self-assembled monolayers (SAMs) are

ordered molecular assemblies with a head group, a spacer and a functional tail

group as represented in Figure 2-10.

Figure 2-10 Schematic diagram of a single crystalline alkanethiol SAM formed on a metal

surface.


SAMs are formed spontaneously when substrates are immersed into a solution

of surface-active molecules in an appropriate solvent [121]. Usually, the head

group is bound to a surface and the tail group presents a chemical functional

group. This group can be modified with drugs or proteins to obtain the desired

functionality such as drug delivery or cytocompatibility. The head and tail

groups are separated by a spacer which is generally an alkyl chain (-CH2). This

organic interface provides well-defined thickness and acts as physical barrier.

Also, crystallisation of this alkyl spacer layer improves monolayer stability.

Hence long spacers are preferred. Commercially available tail group

functionalities range from simple primary amines, hydroxyl and carboxylic

groups to bulky aromatics and epoxides.

SAM coatings are usually nanosized adding only 1-10 nm thickness to the

surface [58]. SAMs can be designed at molecular level to be biologically inert

or active by modifying the tail group with desired functionality. SAMs can be

chemisorbed to numerous materials including glass, metals (gold, copper,

silver, stainless steel, titanium and its alloys, palladium and platinum) and

semiconducting surfaces (silica and indium coated tin oxide) [122].

SAMs coatings are inexpensive and their ease of functionalisation with well-

ordered arrays of SAMs makes them ideal systems for various applications

such as the control of wetting and adhesion, corrosion protection, chemical

sensing, semiconductor passivation, drug/protein delivery and nano-fabrication

[59,122,123]. SAMs have wide applications in the healthcare industry

including blood purification (e.g. algae removal from blood stream using SAM

coated nanoparticles), drug delivery, surface modification of implants and to

study the cellular properties such as interfacial reactions between cells and

organelles and/or proteins and implants [124].

2.5.1. Formation of SAMs

Molecular assembly during the formation of SAMs is a thermodynamic

process. Considering a system at a constant pressure and temperature, the

associated Gibbs free energy (G) can be written as, G = H – TS, where H is the

enthalpy, T is the temperature of the system and S is the entropy. The change

of free energy is given by ∆G = ∆H - T∆S. Self-assembly is driven by the


decrease in Gibbs free energy. For the self-assembly to be spontaneous, the

enthalpy of the system should be negative and in excess of the entropy of the

system. Thus, when ∆G is negative i.e. ∆G = ∆H - T∆S < 0, self-assembly of

molecules take place spontaneously. Minimisation of Gibbs free energy can be

attained by the minimisation of repulsive and/or maximisation of attractive

molecular interactions. Self-assembly process will become progressively less

likely when the magnitude of T∆S approaches the magnitude of ∆H and above

critical temperature [125].

Deposition of monolayers on substrates offer one of the highest quality routes

for preparing chemically and structurally well-defined surfaces [126].

Adsorption of SAMs on to a surface is generally performed either in polar or

non-polar solvents to attain a greater flexibility in the molecular design and

control over the surface properties [59]. SAMs form a covalent bond on

adsorption and the interactions between the subunits (spacers) are non-covalent

(involves variations of electromagnetic interactions between molecules or

within a molecule) such as van der Waals forces. Non-covalent interactions are

critical in maintaining the 3D structure of large molecules. Due to this non-

covalent interaction between the sub-units, the SAMs have the tendency to

shrink and regain their position. SAMs are adsorbed to a surface by various

methods including spray deposition, electrochemical deposition or by simple

immersion [127]. Spatially defined arrays of monolayers can be formed on a

surface by integrating SAMs with patterning methods such as micro-contact

printing and photolithography (a microfabrication technique used to pattern

parts of a thin film or a bulk substrate such as circuits) [115,128].

2.5.2. Assembly of SAMs by simple immersion

The assembly of SAMs by a simple immersion method typically involves the

immersion of the sample into a dilute solution (thiol, silane, phosphonic and

other organic acids), allowing molecules to assemble, then removal of the

sample and rinsing. The attachment and assembly processes of SAMs are

schematically represented in Figure 2-11. The assembly process takes place

once the sample is immersed into the solution. The amphipilic SAM-forming

molecules in solution come in contact with the surface in a few seconds after


immersion leading to the organisation and packing of the molecules to form

well-ordered SAMs in a few hours. As the monolayers continue to form, the

non-covalent interaction between the hydrocarbon chain help in packing

molecules into a well-ordered crystalline layer [129]. Since the head groups

have a special affinity towards metal substrate, formation of multi-layers is not

possible.

Figure 2-11 Steps involved in the adsorption (a) and assembly (b) of SAMs.

The degree of order of these monolayers is mainly determined by the length of

the spacer (alkyl chain) [130]. It has been reported that SAMs with short alkyl

chains (fewer than 10 carbons) lack sufficient attraction between the alkyl

chains to produce well-ordered monolayers. The other important factors that

determine the order of SAMs include cleanliness of the substrate, surface

chemistry and purity of the solvent [58,131]. The important and problematic

contaminants are hydrocarbons (oils which comes from pumps and skin),

impurities on the material’s surface and some polymeric contaminants from the

atmosphere [132]. The quality and assembly of SAMs also depend on several

factors such as head group size, its properties and assembly time. Annealing of

the SAM coated specimens at 120 °C can be performed to obtain a strongly

surface-bound film of monolayers [55]. A detailed explanation on the


assembly of monolayers and how these factors affect the monolayer formation

are described by Schwartz [59].

2.5.3. Types of SAMs

Some of the popular monolayer self-assembly chemistries include:

Thiols, disulfides and thiols on gold, stainless steel [58,123].

Silanes on silicon dioxide and hydroxy functionalised surfaces

[48,133].

Fatty acids and phosph(on)ates on metal oxides [56,134].

Isonitriles on platinum [55].

2.5.3.1. Thiols

An alkanethiol molecule has three chemical entities that determine its

assembly and the general formula is X(CH2)nSH. The head group contains

thiol sulfur (SH) and this acts as the driving force for the chemisorption of

sulfur with the surface. The tail or terminal group (X) can feature one of

several functionalities including a hydroxyl, amine and a carboxylic group.

The terminal hydroxyl (OH) or carboxylic (COOH) groups of the monolayers

are highly useful for the chemical transformations to achieve their desired

functionality. Also, by reacting the carboxylic group of the thiol SAMs with an

acid chloride (SOCl2), further reactions can be attained [123]. For example, if

the terminal group of the monolayer and the reacting group of the binding

molecules have a carboxylic group, then in such cases, the COOH group of the

monolayer can be modified to its corresponding acid chloride and the desired

reaction can be achieved. Scheme 2-1 shows an example of reaction mediated

using thionyl chloride.

Scheme 2-1 Reaction scheme showing the conversion of acetic acid to acetic anhydride using

thionyl chloride


Thiols have been mainly used to modify gold surfaces since 1983 and their

attachment has been well-studied and reported [58,127,131]. However, thiols

have also been used to modify other metallic surfaces including iron and

stainless steel. The alkanethiol monolayers formed on 316L stainless steel (the

letter ‘L’ represents low carbon content of 0.03% max. in the composition)

were not as stable as they are on gold surfaces due to the complex surface

chemistry of 316L SS [58]. Thiols have been used for a variety of applications

including corrosion protection, semi-conductors, biomaterial and biosensing

applications [128,135].

2.5.3.2. Silanes

Similar to thiols on gold, the use of trichlorosilane monolayers on silica

surfaces is popular owing to their excellent grafting properties to silica surfaces

[134]. Silanes are also used in chromatographic techniques for the separation

of molecules. Chlorine in the silane monolayer reacts with the hydroxylated

surface to form a stable siloxane (Si-O-Si) bond. The use of silane monolayers

on metal oxide surfaces such as glass, titanium, aluminium and cobalt-

chromium were also studied in recent years for drug delivery [118,136,137].

Trichlorosilane monomers were observed to be highly reactive towards glass

surfaces [120].

One of the problematic conditions for researchers with silane monolayers is to

synthesise and purify these silane monomers. When the silane monolayers are

assembled on a surface they display a high degree of monomer crosslinking,

condensation and/or formation of multilayers due to siloxane linkages (Si-O-

Si) [138]. Crosslinking was observed to be less stable compared to non-cross

linked monomers and thus allowing the desorption of monolayers in a short

period of time. Mani et al. [137] reported that the silane monolayers formed on

a cobalt-chromium alloy surface were due to the covalent bonding of the

SAMs with the alloy (Si-O-Cr and Si-O-W). However, their study reported

that these monolayers remained ordered and bound to the alloy surface for only

seven days under in vitro conditions [137]. Drug delivery from the implant

surface for a longer period below the toxic level may be necessary for certain

clinical applications. In such cases, this system cannot be used and therefore,


the use of alkoxysilanes instead of chlorosilanes (such as trialkoxysilane) may

be a better option. However, their assembly requires multiple soaking and

annealing steps to form well-ordered monolayers [120].

2.5.3.3. Phosphonate monolayers

Apart from thiols and silane monolayers, phosphonic acids have gained

considerable interest in recent years due to their ability to bond to a range of

metal oxide surfaces and their relatively improved hydrolytic stability under

physiological conditions [129,139]. The use of phosphonic acid SAMs for

surface modification is recent when compared to silanes and thiol SAMs [140].

Similar to thiol on gold, phosphonic acid monolayers adsorb on a metal surface

with a tail-up orientation with a tilt angle (determined using evanescence

reflection spectroscopy) of the hydrocarbon chains of about 30° with respect to

the normal [141].

Phosphonates have three oxygens and a carbon directly attached to a

phosphorous molecule. The binding mechanism of these phosphonic acid

monolayers to metal oxide surfaces can be a monodentate, bidentate or

tridentate as represented in Figure 2-12 [123,142]. This binding mechanism

depends on both the surface and the nature of the organophosphorous

compounds. Phosphonic acids bind to a metal oxide surface by the co-

ordination of phosphoryl oxygen to Lewis acidic sites, followed by the

condensation of P-OH groups with surface hydroxyl group or other surface

oxygen species [143]. A Lewis acid is an electron-pair acceptor and a Lewis

base is an electron pair donor. Thus, the Lewis theory suggests that acids react

with bases to share a pair of electrons, without any change in the oxidation

numbers of the atoms. Phosphonic acid SAMs form a metal phosphonate bond

when adsorbed to a metal oxide surface (Ti-O-P). Adsorption of long chain

alkyl phosphonic acid SAMs on metal oxides has been reported to offer

densely packed and well-ordered SAMs [53,56,144,145]. Like silanes,

phosphonic acid SAMs do not condense (condensation is a reaction in which

two molecules or moieties combine to form a large molecule, together with the

loss of a small molecule) in general due to the absence of homo-condensation


(reaction between the functional groups of the same molecule/moiety) of P-OH

and P-O bonds at aqueous conditions [140].

Figure 2-12 Schematic representation of the binding mechanism of phosphonic acid SAMs to

a metal surface.

Although phosphonic acid SAMs were reported to exhibit high stability, it

depends on various parameters including the density and degree of ordering,

interaction between the monolayers, size of its head group, spacer length and

the reaction method used to chemisorb the monolayers to the surface [146].

Buckholtz et al. [130] attempted to study the effect of long alkyl chains of

phosphonic acid monolayers on Ti6Al4V, titanium, aluminium and vanadium

surfaces. The compared long alkyl chain consisted of carbon atoms in even

numbers between 18 and 30. This study showed that the results were indicative

of the chain length on SAM’s stability. When even numbers of carbon atoms

were used, better packing of the molecules were witnessed than the odd

numbers due to interdigitation (to become interlocked) [147]. Hsu et al. [148]

proposed an air-liquid interface assisted method to form a smooth and a

homogenous phosphonic acid SAMs on a silicon surface. However, the

common method followed by most researchers is the simple immersion

method since it is less complicated and in-expensive.

Tosatti et al. [149] studied the assembly of phosphate monolayers on rough

and smooth titanium oxide coated glass surfaces. Their studies revealed that

phosphonic SAMs were chemisorbed to both rough and smooth titanium

surfaces. The structure and order of phosphonic acid based monolayers on

silicon was studied by Dubey et al. [47] and compared with thiols on gold

surfaces. It was reported that phosphonate monolayers can be used to form


well-ordered SAMs with methyl and hydroxyl groups for electrical and bio-

sensing applications.

The stability of SAMs depends on the temperature, pH of the aqueous medium,

ionic concentrations in the medium and its exposure to harsh environments

such as UV or other high power radiations [123,140]. Bhure et al. [55]

investigated the stability of phosphonic acid monolayers on Co-Cr surfaces and

showed that the monolayers were highly stable when the surfaces were

exposed to atmospheric conditions. In an another study, Kaufmann et al. [56]

investigated the stability of phosphonic acid SAMs on an electropolished Co-

Cr surfaces. This study concluded that the SAMs desorbed in a biphasic

manner in Tris buffer solution with more desorption in the first three days

followed by a slow and sustained release. Their initial fast release was

accounted for by the desorption of physically bound molecules followed by the

sustained release of the covalently bound molecules. An initial study by Mani

et al. [53] showed that the most of the phosphonic acid SAMs adsorbed on a

titanium surface desorbed in a day when immersed in Tris buffer solution at 37

°C. When exposed to ambient air, SAMs were stable for 14 days. However, in

their later studies with an improved SAM adsorption procedure by heat

treating the SAM coated surfaces, stable phosphonic acid SAMs were formed

in solution [37]. Phosphonate monolayers are not only limited to oxide

substrates; they are also used to modify hydroxyapatite and calcium carbonate

[54,150]. However, the number of studies on these non-oxide surfaces is

considerably lower than the oxide surfaces [140].

2.5.4. Biomedical Applications of Phosphonate SAMs

Various surfaces including silicon, titanium, cobalt-chromium, nitinol, gold,

calcium phosphates have been functionalised using phosphonic acid

monolayers for biomedical applications [53,56,145,151]. SAMs can be

functionalised by modifying the tail group with required functionality

including therapeutic drugs and proteins. This can be attained before or after

the adsorption of SAMs to a surface. Although the modifications of SAMs by

both of these processes have been reported, functionalisation of SAMs after its

adsorption to the surface is preferred. This is because, in certain cases, it will


be difficult to control the reactivity of the head group towards the functional

molecules.

There are a number of ways in which the functional group of SAMs can be

modified. SAMs can be adsorbed to a surface and then dip coated with the

active/functional ingredient [144]. In other ways, the functional group of

SAMs are made to react directly with the molecules to be attached. When

performing this, it is possible for both the monolayers and the drug to possess

the same functional groups such as alcohols and carboxylic groups. In these

cases, the SAMs can be converted to their corresponding acid chloride by acid

chloride mediated esterification. Converting the functional group of SAMs to

an acid chloride will enhance the reactivity and will effectively bind the

desired molecule to the SAM. This is mainly because the reactivity of acid

chloride is higher than the hydroxyl or carboxylic group.

Mani et al. [37] reported the immobilisation of flufenamic acid (as a model

drug) on to gold and titanium surfaces using SAMs with carboxyl group. Thiol

SAMs were used for gold whereas phosphonic acids SAMs were used for

titanium. In their study, acid chloride esterification (ACE), dry heat

esterification (DHE) and direct esterification (DE) reactions were performed to

immobilise flufenamic acid to these surfaces. Briefly, for the ACE, 1.125 g of

the drug was refluxed with 4 mL of thionyl chloride (SOCl2) in an inert

atmosphere. After an hour, the excess thionyl chloride was evaporated and the

drug was dissolved in a solvent (THF). Sample specimens of both phosphonic

SAMs on titanium and thiol SAMs on gold surfaces were immersed in this

solution and incubated for 48 hours. In the DHE, both thiol and phosphonic

acid SAM coated surfaces were covered with 200 µg of flufenamic acid. The

specimens were then heated to 150 °C in air at a constant temperature for 4

hours. In the DE procedure, a stock solution of 525 µg of flufenamic acid in

105 mL of anhydrous toluene was prepared. SAM coated samples were

immersed in 10 mL (for each sample) of the stock solution. Finally, the

samples treated with the three esterification methods were rinsed with solvents

and deionised water for several minutes to remove physisorbed molecules.


Surface characterisation by X-ray photoelectron spectroscopy showed the

successful attachment of flufenamic acid to both thiol and phosphonic acid

SAM coated surfaces. The total amount of drug loaded to the titanium surface

was 0.93 ± 0.24 ng/mm2 (DHE) and 0.61 ± 0.11 (DE) compared to 1.27 ± 0.45

ng/mm2 (DHE) and 0.13 ± 0.02 ng/mm

2 (DE) for gold surface. According to

Mani et al. [37], over all, approximately 0.1 – 1 ng/mm2 was coated. The drug

was observed to release in a biphasic manner i.e. initial burst release and a

sustained release. For both DHE and DE, nearly 50% of the drug was released

by day one, followed by a sustained release of the remaining drug.

The results for ACE were not presented in their study as they found significant

scattering of the results for these samples which may be due to various factors

as discussed below. Generally, an acid chloride will readily react with

atmospheric moisture/humidity. The authors performed refluxing of the drug

with thionyl chloride in an inert atmosphere to form an intermediate compound

with an acid chloride group. However, after an hour, the excess of thionyl

chloride was evaporated by exposing the refluxed mixture to atmosphere.

During this evaporation stage, since the mixture was exposed to air, the acid

chloride in the intermediate compound will react with the moisture to form

flufenamic acid again (Scheme 2-2). Also, the THF used to dissolve the

intermediate compound and immerse the samples was not reported to be

anhydrous. Hence, the intermediate compound with acid chloride group would

have reacted with any water present in the THF. Thus, these reactions would

have converted the intermediate compound (refer to scheme 2-2) back into the

drug, flufenamic acid.

Atomic force microscopy (AFM) images showed the high possibility for

sedimentation and/or crystallisation of the drug on to the surface for most DHE

and DE samples (Figure 2-13). This can be backed up with the initial burst

release of more than 50% of the drug within day 1 since cleavage/hydrolysis of

the drug from the monolayer is generally expected to be a slow process at a pH

of 7.4. Thus, this study did not provide a suitable method to immobilise

flufenamic acid and the results were inconclusive.


Scheme 2-2 Scheme of reactions on acid chloride esterification of flufenamic acid. Label

description: Flufenamic acid reacting with thionyl chloride (1); reaction intermediate (2);

reaction intermediate reacting with water (3); flufenamic acid (4).

Figure 2-13 Atomic force microscope images of flufenamic acid functionalised on gold

surface using acid chloride esterification (a) and flufenamic acid coated to a titanium surface

by dry heat esterification method (b). Reproduced with kind permission from Elsevier [37].

Their later studies on immobilising an anti-cancer drug Paclitaxel™ on a

cobalt-chromium metal surface with and without SAMs showed some

interesting results [45,144]. Their method to bind Paclitaxel™ to SAM coated

surfaces has a number of limitations. In their procedure, a microdrop method

was used to drop the drug solution onto the cobalt-chromium surface. It was

reported that from the 25 µg of drug placed on the surface (1 cm2), only 25%

of the drug was adsorbed to the surface and the rest washed away by solvent

cleaning. Although a certain amount of the drug can be recovered by recycling,

this method can be time-consuming and expensive. Also it should be noted that

scaling up of this method will be difficult since coating drugs on implants with

complex geometries using microdrop deposition would be challenging.


Furthermore, the drug adsorbed on the surface was observed to be crystallised.

Their study on the delivery of Paclitaxel™ from the monolayers showed a

sustained release; however, similar to their previous studies, crystals of

Paclitaxel™ were formed. Crystallisation of drugs on an implant surface can

potentially increase the surface roughness of the implant and may cause post-

implant complications. Also due to crystallisation, controlling drug release

from the implant surface will be problematic.

Torres et al. [54] reported the use of phosphonic acid monolayers to deliver an

antibacterial drug from a hydroxyapatite surface to prevent bacterial infections

and bio-film formation. Silver has been previously reported for its antibacterial

activity [46]. In this study, phosphonic acid SAM coated surfaces were

immersed in silver nitrate solution to functionalise the SAMs with silver. A

significant reduction of bacterial adhesion onto the functionalised

hydroxyapatite surface was observed. Apart from a therapeutic drug,

functionalisation of a protein (RGD peptide) to phosphonate SAMs adsorbed

on pure titanium and Ti6Al4V surfaces for osteoconductive surfaces was

demonstrated by Schwartz et al. [152]. Raman et al. [153] used hydrophobic

phosphonate SAMs to control the non-specific fibroblast cell adhesion and

proteins on to stainless steel substrates. Using SAM patterning methods such

as lithography, immobilisation of monolayers with proteins and cells to form

patterned surfaces have also been explored [154].

2.6. Knowledge Gap

2.6.1. SLM

Titanium and its alloys including Ti6Al4V, Ti6Al4V ELI (extra low

interstitial) and Ti6Al7Nb are commercially used to build implants for

orthopaedic, dental and vascular applications due to their excellent corrosion

resistance, mechanical properties, high strength-to-weight ratio and inertness

[109,155]. A thin and stable passive titanium oxide (TiO2) layer (3-10 nm

range) is formed when the surface is exposed to ambient air or water and

contributes to the chemical inertness of titanium and its alloys [156]. Due to

the ability of Ti and its alloys to form this passive and stable oxide layer, it has

wide application in harsh and corrosive environments [157]. The stability of


this surface oxide layer can be influenced by several factors including

crystalline structure and thickness; however, surface chemistry plays a pivotal

role [155].

In the SLM process, the powder material is melted and cooled rapidly. Hence,

there is a high possibility for an SLM fabricated part to differ in its surface

chemistry and microstructure compared to conventionally manufactured parts

(e.g. those made by casting and forging). A selective remelting strategy

proposed in the literature [93] may benefit the part to improve its density,

surface morphology and mechanical property but at the same time, it may lead

to varied surface chemistry and distribution of the alloying element within the

part. When a high amount of heat is applied especially to an alloy (by

remelting of the layers), the alloy may be superheated and will reach a high

temperature. Precipitation of the alloyed elements is possible if the alloy

remains at such high temperature for a long time. For example, it has been

shown that segregation of Al can occur during the SLM of Ti6Al4V due to

rapid melting and cooling of the alloy [158]. Such segregation of alloying

elements can potentially alter the surface chemistry significantly by altering

the surface oxide composition leading to surface heterogeneity.

Surface heterogeneity due to the impact of the laser can have a significant

impact on biomedical applications. A biomaterial interacts with the biological

environment at a bio-interface. Bio-interphases are affected by the nature of

the biomaterials including its surface energy, surface chemistry, surface

morphology and surface roughness [1,4,13]. Without a good understanding of

the surface chemistry, the interaction of cells at the interfaces will be poorly

understood.

Surface chemistry of a part mainly depends on the distribution of its alloying

elements on the surface and the atmosphere to which the part is exposed to.

Inclusion and discontinuity spots that arise due to surface heterogeneity may

affect the biocompatibility [159]. For example, in the case of Ti6Al4V, the

presence of aluminium or vanadium on the surface may form their

corresponding oxides and reduce the stability of the surface oxide layer which

can affect the cytocompatibility that may lead to cytotoxicity and poor cell


adhesion under in vivo conditions [3,132]. The presence of alloying elements

on the implant surface can influence the dissolution of ions in to the biological

environment. Clinical experience has shown the release of vanadium ions from

implants made of Ti6Al4V to the body [160]. These metal ions interfere with

cell differentiation and contribute to periprosthetic osteolysis by impairing

osteogenesis [161]. Hence surface modification of Ti6Al4V components or the

adoption of a different biocompatible material is required.

Although several surface modification techniques such as passivation,

chemical etching, ion-implantation and chemical coatings are available to

improve and modify the surface chemistry of an implant [1,4,117], the

principle source of those surfaces is the manufacturing process. The knowledge

on the surface chemistry of a SLM fabricated part and the impact of laser

power on its surface chemistry is limited. Hence a study on the surface

chemistry of SLM fabricated parts is central for a better understanding of its

biocompatibility and the possible risk associated with the use of biomedical

implants fabricated using SLM into the human body.

2.6.2. SAMs

Although a few studies have reported on the assembly of monolayers to rough

and smooth surfaces, there is little literature that reports on the chemisorption

and stability of phosphonic acid SAMs to highly porous and rough surfaces.

As SAMs can serve as a localised drug delivery system, the use of SAMs to

attach drugs to biomedical implants is gaining considerable interest. Although

the delivery of drugs using phosphonic acid SAMs have been achieved by a

few researchers, most of these drugs are either physisorbed or both

physisorbed and covalently bound leading to a burst release. In situations

where a sustained release is preferred, burst release can significantly affect the

requirement. One-to-one chemical binding of a drug with a monolayer can

control and prevent the initial burst release of the drug. Although the acid

chloride esterification method used by Mani et al. [37] is inconclusive due to

their exposure to atmospheric air, further optimisation of this method may

potentially render the required binding of drugs to monolayers.


It may be difficult to functionalise SAMs with a drug using a micro-drop

method where the geometry of the part is complex. Using the micro drop

method in such cases may not render a homogenous coating. For example, the

use of micro drop method to coat drugs to cardiovascular stenting applications

will be highly complicated since their typical strut thickness is only 100 µm.

Also, the use of this method for porous implants with complex internal and

external geometries would be difficult. Hence a simple and efficient method to

functionalise drugs to monolayers is required.

Most previous literature has discussed the coating methods, attachment,

stability and release profile of the immobilised drug to SAMs. However, the

knowledge on the activity of the drug upon its delivery from SAMs is limited.

For example, instead of the drug, if the monolayer desorbs from the surface

together with the drug, this will not produce the required activity. Hence it is

important to study if the functionalised drug is active upon its release from the

monolayer.

In addition to these current knowledge gaps, according to the best knowledge

of the author, the use of phosphonic acid monolayers to functionalise SLM

fabricated Ti6Al4V surfaces with therapeutic drugs has not been reported.

2.7. Summary

The selective laser melting process has been identified as one of the best routes

to fabricate customised/patient-fitting biomedical devices. The versatility of

phosphonic acid monolayers to form densely packed, well-ordered and highly

stable SAMs on metal oxide surfaces makes them ideal for these biomedical

applications. Delivery of drugs from monolayers attached to metal surfaces

produced by various conventional manufacturing methods has previously been

studied. However, to date, there have not been reports in the literature that has

explored the possibility to attach monolayers to a SLM fabricated surface to

the best of the author’s knowledge. Hence research on integrating these two

processes is required to address the knowledge gap and bridge these

technologies to fabricate customised implants with drug delivery options.

Chapter 3: Research Hypothesis and Novelty 55

3. RESEARCH HYPOTHESIS AND NOVELTY

3.1. Aim of the Research

To evaluate the potential to combine selective laser melting (SLM) to fabricate

customised implants with surface modification using self-assembled

monolayers (SAMs) for precise drug delivery.

3.2. Objectives

To study the surface morphology and surface chemistry of Ti6Al4V

components fabricated by SLM.

To coat phosphonic acid SAMs onto SLM fabricated surfaces.

To investigate the in vitro stability of SAMs on the SLM as-fabricated

and mechanically polished surfaces.

To develop an approach to functionalise the SAMs adsorbed on a SLM

fabricated surface with therapeutically active substances.

To determine the amount of drug coated and the stability (including

oxidative exposure and in vitro) and biological activity of the

immobilised drug molecule.

3.3. Research Novelty and Contribution to Existing

Knowledge

This research presents a novel multi-disciplinary approach to fabricate and

functionalise complex biomedical implants by integrating SLM with surface

modification using SAMs for precise drug/protein delivery to improve

mechanical and biological properties.

Although there is significant understanding of SLM process conditions and

process parameters, there is very limited awareness of the surface chemistry of

the SLM fabricated surfaces. By exploring the surface chemistry of a SLM

fabricated Ti6Al4V surface and comparing it to a conventional surface, this

research contributes to furthering the knowledge of this AM process. Currently

there is no literature on the surface modification of SLM fabricated Ti6Al4V

parts in terms of altering its surface chemistry using phosphonic acid SAMs to

Chapter 3: Research Hypothesis and Novelty 56

deliver therapeutics on-site. This will be the first study according to the best

knowledge of the author. The process-optimised acid chloride esterification

reaction used in this study to immobilise therapeutic drugs to phosphonic acid

SAMs will contribute to the knowledge available for functionalising SAMs

with drugs. Previous studies on immobilisation of a drug to monolayers did not

show the therapeutic activity of the released drug from the monolayers. In this

thesis, the activity of the drug is examined.

Chapter 4: Research Methodology 57

4. RESEARCH METHODOLOGY

4.1. Summary of Experimental Procedure

Initially tracheobronchial stents (cylindrical meshes used to maintain airway

patency in trachea and bronchi) were fabricated using SLM and surface

modified using SAMs. Stents were successfully fabricated from 316L stainless

steel (316L SS), one of the materials used commercially to produce stents.

Although SLM was capable of fabricating stents of 300 µm strut thickness, the

SLM-produced stents were rough due to partial sintering of particles to the

struts. Electropolishing was performed to remove the sintered particles and

mechanical testing was performed to determine the compressive strength of the

SLM fabricated stents. Electropolishing yielded a relatively smooth surface,

however, it removed nearly 30% of the material from the stent. On mechanical

testing, the stents deformed plastically. From the results, it was identified that

the SLM as-fabricated stents did not possess the required mechanical and

surface properties and needed more post-processing than the conventionally

produced stents using laser cutting. Also, coating of monolayers to this thin

stent struts and characterisation of the stent struts using X-ray photoelectron

spectroscopy (XPS) and surface wettability measurements were not

straightforward. Hence, it was decided to perform research on the attachment

of SAMs to the SLM fabricated cuboid surfaces and functionalise the SAMs

attached to these surfaces using therapeutic agents. A detailed report on the

fabrication and mechanical testing of tracheobronchial stents using SLM is

annexed in appendix 1.

Later, an experiment was performed to surface modify SLM fabricated 316L

SS cuboid samples using thiol monolayers. Consistent with previous literature

on SAM modification of 316L SS surfaces, the adsorbed SAMs were not

stable. Studies conducted on the available monolayers to modify metal oxide

surfaces revealed phosphonic acid SAMs as one of the best candidates for the

purpose. To examine this, cuboidal samples were fabricated in a SLM machine

using three commonly used metallic biomaterials 316L SS, Ti6Al4V (titanium

alloy) and L605 cobalt-chromium alloy. Surface modification of these as-


fabricated surfaces was performed using phosphonic acid monolayers. XPS

characterisation confirmed the adsorption of SAMs to all three surfaces

(Appendix 2). Since phosphonic acid SAMs showed the covalent attachment to

all three metal oxide surfaces, it was decided that the use of all these three

surfaces to address the aim was not essential. Also, studies on all three surfaces

would increase the number of samples to be characterised and would be time

consuming. Hence, to narrow down the research, one material was selected as

the candidate member to address the aim.

Ti6Al4V, a grade 5 titanium alloy, is a material that can be processed relatively

easily by SLM and is being researched extensively for biomedical applications

[14,15,162]. Titanium and its alloys are ranked the best for corrosion resistance

(due to its stable surface oxide layer), than cobalt, nickel titanium alloys and

stainless steel [111]. It is also preferred for load bearing application due to its

excellent strength-to-weight ratio compared to 316L SS and Co-Cr alloy. In

addition to this, previous studies reported the successful adsorption of

phosphonic acid SAMs to Ti6Al4V surfaces fabricated by conventional

methods. Hence, Ti6Al4V was selected as the material for this research.

Figure 4-1 shows a schematic of the research methodology used in this study.

To begin with, the surface chemistry of Ti6Al4V virgin powder and recycled

powder (in an AM 250 SLM machine) was examined using XPS to determine

any changes in the surface chemistry of the virgin powder. Since no significant

change was observed, cuboid samples were then fabricated using the recycled

Ti6Al4V in the AM 250 machine (previously SLM 250). The surface

chemistry of the SLM as-fabricated (SLM-AF) and mechanically polished

(SLM-MP) samples were studied and compared with a conventionally forged

Ti6Al4V sample. The impact of laser power and selective laser re-melting on

the surface chemistry of SLM fabricated samples was also studied.


Figure 4-1 Schematic of the research methodology

A set of SLM fabricated samples were polished mechanically and surface

cleaned to remove contaminants. Phosphonic acid SAMs were used to modify

SLM fabricated Ti6Al4V parts due to their versatility and increased stability

on metal oxide surfaces. Carboxylic terminated phosphonic acid SAMs were

adsorbed to the SLM-AF and SLM-MP samples. In vitro stability of the

carboxylic acid terminated SAMs on both SLM-AF and SLM-MP surfaces

were investigated in Tris-HCl buffer solution of pH 7.4. On studying the in

vitro stability, mechanically polished samples were used for drug attachment

and other further studies. This is because, the electron counts for SLM-MP was

better than SLM-AF. The presence of partially melted particles on the SLM-

AF surface was observed to affect the number of electrons reaching the

detector and thus reducing the signal.

Paracetamol was initially used as the model drug to functionalise the SAMs,

followed later by Ciprofloxacin®. On successful attachment of

Ciprofloxacin®, quantification of the drug immobilised, oxidative and in vitro

stabilities, drug release profile and antibacterial activity of the immobilised

drug were determined.


4.1.1. Materials

The materials and chemicals used during this research are tabulated in Table 4-

1.

Table 4-1 List of materials/chemicals used.

Material/Chemical Name Chemical Formula Supplier

Purity/

Grade

Titanium alloy powder (Ti6Al4V) -

LPW

Technology

Ltd., UK.

ASTM

F136

Hot forged Ti6Al4V plate -

TIMET UK

Ltd., UK. N/A

Silicon carbide grits - Buehler, UK. N/A

Diamond pastes (6 µm and 1 µm) - Buehler, UK. N/A

Extender solution - Buehler, UK. N/A

Dichloromethane (DCM) CH2Cl2

Sigma

Aldrich, UK. 99.5%

Methanol CH3OH

Sigma

Aldrich, UK. 99.8%

Ethanol C2H5OH

Sigma

Aldrich, UK. > 99.5%

Hydrogen peroxide H2O2

Fisher Scien-

tific, UK 30%

Sulfuric acid H2SO4

Sigma

Aldrich, UK.

95 –

98%

Hydrochloric acid HCl

Sigma

Aldrich, UK. 37%

Sodium hydroxide NaOH

Sigma

Aldrich, UK. ≥ 98%

Thionyl Chloride SOCl2

Sigma

Aldrich, UK. > 99%

Tetrahydrofuran (THF) C4H8O

Sigma

Aldrich, UK. ≥ 99.9%

16-Phosphanohexadecanoic acid (16-

PhDA) HOOC(CH2)15PO2H

Sigma

Aldrich, UK. 97%

Octadecylphosphonic acid (ODPA) C18H39O3P

Sigma

Aldrich, UK 97%

Tris(hydroxymethyl)aminomethane NH2C(CH2OH)3

Sigma

Aldrich, UK ≥ 99.8%

Paracetamol C8H9NO2

Sigma


Ciprofloxacin® C17H18FN3O3

Sigma


Carbon film/tape -

Agar

Scientific,

UK. N/A

Muller-Hinton Agar - Sigma

Aldrich, UK.

For

micro-

biology

Sterile filter paper disc 10 mm diameter -

Sigma

Aldrich, UK.

Micro-

biology


4.1.2. Equipment

4.1.2.1. Selective Laser Melting

Ti6Al4V components were fabricated in a Renishaw AM 250. The machine

was equipped with a fibre modulated pulse laser with a maximum power of

200 W and wavelength (λ) of 1070 nm. The AM 250 machine consists of a

hopper, a wiper, an elevator that lowers the substrate to adjust the layer

thickness and a lens that focuses the laser (200 W maximum) to the build area

(250 x 250 x 300 mm). Before SLM, the Ti6Al4V powder from which the part

was to be fabricated was spread over the build platform from the hopper to a

pre-defined layer thickness. After the layer had been spread, the laser beam

scanned and fused the powder in the areas specified by the layer of the CAD

file.

4.1.2.2. Malvern Mastersizer

Particle size distributions (PSDs) were measured using a Malvern Mastersizer

3000 (Malvern, Worchestershire, UK) by adopting a dry dispersion method.

This equipment works on the principle of laser diffraction. Laser diffraction

measures PSDs by measuring the angular variation in the intensity of light

scattered as a laser beam passes through a dispensed particulate sample. This

angular scattering intensity data is then analysed to calculate the size of the

particles responsible for creating the scattering pattern. To measure the PSD,

the Ti6Al4V powder feed rate was set to 25% and the dispersion pressure was

1 bar. PSDs were obtained for five sets of sample dispersion and averaged.

Built-in software was used to calculate the PSDs.

4.1.2.3. Scanning Electron Microscope

A Philips (PW 6800/70) scanning electron microscope was used to obtain the

surface morphology of the SLM fabricated Ti6Al4V surface. The scanning

electron microscope (SEM) was operated at a voltage of 20 kV, spot size of 10

nm and morphologies of SLM-fabricated surfaces were obtained at various

magnifications such as 100x, 250x and 500x.


4.1.2.4. Surface Profilometer

Surface roughness (Ra) patterns of the SLM as-fabricated samples and

mechanically polished samples were obtained using an Alicona InfiniteFocus®

optical 3D measurement device. The Ra value was calculated by averaging the

values obtained for five different samples from 175 µm2 of each sample.

4.1.2.5. Contact Angle Goniometer

The surface wettability of the sample was characterised using contact angle

goniometry. Static water contact angle measurements were performed using an

OCA 20 contact angle goniometer (DataPhysics, Filderstadt, Germany) at

ambient temperature. A drop volume of 5 µL deionised water (pH ~ 6.8) was

placed on an arbitrary spot of the surface and allowed to settle for five seconds.

Images were acquired at standard atmospheric temperature and pressure using

the inbuilt software and the contact angles were measured. The static water

contact angles have been reported in this work as a mean ± standard error of

the mean of three distinct spots on five different samples.

4.1.2.6. X-ray photoelectron spectroscopy

The surface chemistry of the samples was examined using X-ray photoelectron

spectroscopy (XPS), also called electron spectroscopy for chemical analysis

(ESCA). An XPS is built with an X-ray source to produce photoelectrons, a

sample stage to place the sample, a lens, an ion gun to etch the sample, a

detector to collect the electrons and an analyser to produce the result [163].

XPS analysis was performed at a high vacuum pressure of 10-8

Pa so that the

electrons can travel to the detector. Monochromators were used to increase the

energy resolution and to eliminate satellite signals from Kα3 and Kα4 X-rays.

In XPS, the sample is irradiated with soft X-rays. Kinetic energy (Ek) of the

emitted electrons due to irradiation is measured. The binding energy (EB) of

the photoelectrons can be calculated using the equation EB = hν – Ek – φ,

where, hν is the energy of the X-ray photon and φ is the spectrometer work

function. The most commonly used X-ray sources are Al Kα (1486.6 eV) and

Mg Kα (1253.6 eV). XPS can determine the surface chemistry of a sample to a


depth of the first 10 nm and is considered as a vital tool for surface chemical

analysis [164,165].

In this research, a Thermo Scientific K-Alpha XPS was used to obtain the

surface chemistry of the Ti6Al4V surface. Depth profiling was performed to

study the elemental composition and the oxide layer thickness of the SLM

fabricated Ti6Al4V surface. Depth profiles were obtained by sputtering the

specimen at a rate of approximately 1.35 Ås-1

(for titanium) using an argon ion

gun. The probing spot size used to obtain the measurement was 400 µm.

Aluminium (Al) Kα monochromated radiation at 1486.6 eV was used and

photoelectrons were collected at a take-off angle of 90°. Survey spectra were

collected at a pass energy of 100 eV in a constant energy analyser mode. High

resolution spectra were obtained at a pass energy of 20 eV. Peak deconvolution

was performed using Gaussian-Lorentzian curves and was used to investigate

the different chemical states. A built-in Thermo Avantage data system was

used for data acquisition and processing. The national institute of standards

and technology (NIST) database and previously reported literature were used

to identify the unknown spectral lines. The thickness of the oxide film was

estimated by two methods. In method 1, the thickness was calculated from the

depth at which the oxygen content decreased to half of its maximum value

during depth profiling. In method 2, the thickness was estimated from the

depth at which the major alloying element Ti intersects the O in the depth

profile [166].

The possible errors in the atomic percentage due to peak fitting uncertainties

were calculated using CasaXPS software. The software calculates the standard

deviation of a peak using a Monte Carlo method [167]. During this

calculation, fitting trials are conducted in which synthetic noise is added to the

measured data, followed by a peak fit. This analysis was performed for all

detected peaks in a survey spectrum. The standard deviation of the fitted areas

is used as the area error. Fractional area error was calculated as the ratio

between the actual area of the peak to the area error. This fractional area error

was multiplied with the atomic percentage of detected elements to obtain the

final error. The estimated error percentage was between 0.02 and 0.11 for


different elements (i.e. 2 – 11%). Henss et al. [168] also accounted the XPS

quantification error as ± 10%. Thus the instrumental error due to calibration for

this study was taken as ± 10% in addition to the reported mean ± standard error

of the mean.

4.1.2.7. Fourier Transform Infrared Spectroscopy

Perkin Elmer Spectrum 100 FTIR was used to characterise the Ciprofloxacin®

coated on the SLM fabricated Ti6Al4V substrates (Figure 4-2). Initially, the

crystal area of the Spectrum 100 FTIR was cleaned with soft tissue dipped

with ethanol. The FTIR was operated under Attenuated Total Reflectance

(ATR) mode. A background scan was collected and the cleanliness of the

diamond stage was ensured. The Ti6Al4V substrate whose surface is to be

characterised is placed on the diamond stage. Once the sample was placed, a

pressure arm in the instrument was screwed down on to the sample placed over

the prism (diamond) stage. The pressure arm was locked into the precise

position above the diamond crystal. The pressure arm applies a small amount

of force (~ 80 N) to the sample, pushing the sample on to the diamond surface

using the pressure arm. This pressure ensures a good contact between the

sample surface and the diamond stage, and prevents the sample from moving

during the characterisation. A force gauge is present in the instrument to warn

when too much of pressure is applied to the stage. A control Ti6Al4Vsurface

without any drug coating was characterised and used as background for the

drug coated samples. All samples were analysed from 4000 cm-1

to 380 cm-1

with a spectral resolution of 4 cm-1

.

Figure 4-2 Spectrum 100 Fourier Transform Infrared spectrometer fitted with a pressure arm

(with the kind permission from Perkin Elmer).


4.1.2.8. UV-Vis spectrophotometers

UV-Vis spectrophotometers are used to measure the absorbance of UV or

visible light by a sample either at a single wavelength or perform a scan over a

range in the spectrum. UV-Vis spectroscopy is the measurement of light when

it passes through a sample and its principle is based on the ability of a

molecule to absorb ultraviolet and visible light. This technique can be used

both quantitatively and qualitatively. A schematic diagram of a UV-visible

spectrometer is shown above in Figure 4-3.

Figure 4-3 Schematic of a UV-Visual Spectrophotometer

The UV region ranges from 190 to 400 nm and the visible region from 400 to

800 nm. The light source (a combination of tungsten/halogen and deuterium

lamps) provides the visible and near ultraviolet radiation covering the 200 –

800 nm region. The output from the light source is focused onto the diffraction

grating which splits the incoming light into its component colours of different

wavelengths, like a prism (shown above in Figure 4-3) but more efficiently.

For liquids, the sample is held in an optically flat, transparent container called

a cell or cuvette. The reference cell or cuvette contains the solvent/solution in

which the sample is dissolved and this is commonly referred to as the blank.

The sample cell contains the solution whose concentration is to be measured.

For each wavelength the intensity of light passing through both a reference cell

(Io) and the sample cell (I) is measured. If I is less than Io, then the sample has

absorbed some of the light. The absorbance (A) of the sample is related to I

and Io according to the following equation:

A = Log10 l0/l


The detector converts the incoming light into a current, the higher the current

the greater the intensity. Then the absorbance is plotted against wavelength

(nm) in the UV and visible section of the electromagnetic spectrum.

According to Beer’s law, absorption is proportional to the number of absorbing

molecules and Lambert’s law tells us that the fraction of radiation absorbed is

independent of the intensity of the radiation. Combining these two laws,

𝑨 = 𝜺𝒄𝒍

where ‘A’ is the absorbance, ‘ε’ is the molar absorptivity and expressed in

units L mol-1 cm-1

, ‘C’ is the concentration of the sample (compound) and

expressed as mol L-1

and ‘l’ is length of cell and expressed in units cm. If the

absorbance of a series of sample solutions of known concentrations are

measured and plotted against their corresponding concentrations, the plot of

absorbance versus concentration should be linear if the Beer-Lambert Law is

obeyed. This graph is known as a calibration graph. A calibration graph can be

used to determine the concentration of an unknown sample solution by

measuring its absorbance.

4.1.3. Methods

4.1.3.1. CAD Model

Sample parts with dimensions of 10 mm x 10 mm x 3 mm were designed using

Magics 14.1 (Materialise) software and saved in STL file format. This design

was then replicated to fabricate the required number of samples. Figure 4-4

represents the build platform with replicated models (in red) and the meander

laser scan strategy used (right-hand side). For antibacterial susceptibility

testing, cuboid samples of 25 x 30 x 2 mm were designed using Magics.


Figure 4-4 Representation of sample models aligned in the build platform and the Meander

scan strategy used for fabrication (right hand corner)

4.1.3.2. Sample Fabrication

Parts were fabricated using a SLM machine (AM 250). Previously optimised

SLM process parameters used for the study are tabulated in Table 4-2. A multi-

directional meander scan strategy (Figure 4-3) was used where the laser scan

direction was rotated by 67° for each layer to reduce the residual stress. The

components were built on a Ti6Al4V metal substrate preheated to 80 °C. The

build chamber was filled with argon gas to maintain an inert atmosphere. The

temperature of the build chamber during the process was measured to be 34 –

36 °C. Parts were fabricated with skin scan (re-melting of the top surface using

laser to improve the surface quality) for a better surface quality. However, to

study the effect of skin scan on the surface chemistry, parts without skin scan

were also fabricated. Apart from this, all other parameters were constant. The

SLM fabricated samples were removed from the substrate upon fabrication.

Table 4-2 Summarised SLM Process Parameters

Laser Power P (W) 200

Hatch Spacing (µm) 100

Point Distance (µm) 50

Exposure Time (µs) 220

Layer Thickness (µm) 50

Scan Strategy Meander

Build Plate Temperature (°C) 80


4.1.3.3. Sample Preparation

The SLM fabricated samples were sonicated in dichloromethane (DCM),

methanol, ethanol and deionised water for 10 minutes each to remove loosely

bound Ti6Al4V particles from the sample surface and contaminants that might

be on the surface. The substrates were then dried using compressed air. The

cuboid samples with 10 mm x 10 mm x 3 mm dimensions were used for the

surface chemical, surface modification and functionalisation studies. For

antibacterial susceptibility testing, discs of 8 mm diameter were laser cut from

the cuboid samples (25 mm x 30 mm x 2 mm) fabricated by SLM.

4.1.3.4. Mechanical Polishing

The rinsed and dried samples were mechanically polished on only one side (10

mm x 10 mm) using a series of silicon carbide grits (P200, P400, P600, P800

and P1200) for 5 minutes each. These samples were then polished using

diamond paste of 6 µm and 1 µm for 3 minutes each. Mechanical polishing

was performed to make the surface smooth for characterisation purposes and to

study the surface chemistry of a mechanically polished SLM fabricated

surface. It should be noted that the cuboid samples were hand held while

polishing and not clamped (difficult to hold the sample in place while

polishing) or mounted using any resins. This is because the use of resins may

affect the surface chemistry of the sample and may lead to contamination. The

polishing fluid may also contribute to a certain level of contamination to the

surface. This is cleaned using solvents and acids as described in the sample

cleaning section below. Also, removal of the sample from the mount can be

tedious. The discs for the antibacterial susceptibility testing were also polished

in the same manner i.e. hand held. Samples polished using the above

mentioned procedure rendered the surface roughness of 590 ± 13.5 nm.

4.1.3.5. Surface Cleaning

The purity and cleanliness of a sample’s surface plays an important role in the

assembly of SAMs since the assembly of monolayers are affected by the

presence of contaminants [58,131]. In order to attain well-ordered monolayers,

freshly prepared surfaces are recommended. Hence cleaning of SLM fabricated


samples was performed to ensure the cleanliness and purity of the surfaces

before assembling monolayers.

Several cleaning techniques have been reported in the literature for assembling

monolayers. A mixture of solution was prepared with concentrated sulfuric

acid, hydrogen peroxide (30%) and water in the ratio of 1:1:5. Sulfuric acid

was added to water followed by hydrogen peroxide. Addition of water or

hydrogen peroxide to Sulfuric acid is exothermic and the resultant heat may

increase the sample temperature above 100 °C. Hence, care should be taken.

The samples were immersed in this solution for 15 minutes to remove any

surface contaminants. The samples were then rinsed in deionised water for 30

minutes twice to remove any residues of the chemical mixture. The samples

cleaned using this technique rendered a highly wettable surface (21° ± 0.9°)

compared to solvent cleaning mentioned in the sample preparation section.

Hence, this method was used throughout the study.

4.1.3.6. Preparation and adsorption of SAMs

Two types of phosphonic acid SAMs including a carboxylic acid terminated

16-phosphanohexadecanoic acid (16-PhDA; SAMs) and a methyl terminated

octadecylphosphonic acid (ODPA) SAMs were used. 1mmols of 16-PhDA and

ODPA SAMs were prepared separately in dry THF and coated on the Ti6Al4V

surface using a solution immersion deposition method previously reported

[56]. Briefly, the cleaned samples were rinsed with THF and immersed in the

SAM solutions immediately. After 24 hours, the samples were removed from

the THF and the residual SAM solution on the surfaces was allowed to

evaporate in air. The samples were then transferred to an oven at 120 °C

without rinsing. After 24 hours, the samples were removed from the oven and

allowed to cool at room temperature. Finally, the samples were sonicated in

THF and deionised water for 1 minute each to remove any physisorbed SAMs.

4.1.3.7. Functionalisation of SAMs with Ciprofloxacin®

Both 16-PhDA and ODPA coated SAMs were functionalised with

Ciprofloxacin® separately to avoid cross contamination. As ODPA has a

methyl terminal group, Ciprofloxacin® cannot be attached to the terminal end


of these monolayers whereas 16-PhDA has a carboxylic group in its terminal

end that can be functionalised. Hence, the ODPA coated surfaces were used as

a control sample to prove the surface reaction of Ciprofloxacin®. The

apparatus include a round bottomed flask (RBF) and a reflux condenser with a

constant flow of nitrogen gas into the RBF to perform the reaction under an

inert atmosphere. The apparatus was dried using a flame/heat gun to avoid the

presence of moisture. A sonicator was used to agitate the sample. The reaction

steps that followed to coat and functionalise the 16-PhDA SAMs with

Ciprofloxacin® are shown in Scheme 4-1.

Scheme 4-1 General scheme of reactions to coat 16-PhDA monolayers to Ti6Al4V surface and

to functionalise the monolayers with Ciprofloxacin®. Label description: Hydroxylated

titanium surface (1), 16-Phosphanohexadecanoic acid (2), SAMs adsorbed on titanium surface

(3), reaction intermediate formed as a result of acid chloride esterification (4) and

Ciprofloxacin® immobilised to SAMs.


Firstly, the 16-PhDA SAM coated Ti6Al4V surfaces were transferred to the

RBF with the SAM coated surface facing upwards. 3 mM of thionyl chloride

(SOCl2) in anhydrous THF was added to the RBF and the apparatus was

sonicated for an hour. Care was taken to ensure that the solution covered the

samples to obtain a uniform reaction. After one hour, the solution in the RBF

was syringed out and the Ti6Al4V surfaces were sonicated thrice for 3 minutes

by adding an excess of anhydrous THF to the RBF each time. After this step,

10 mM of Ciprofloxacin® in anhydrous THF was added to the RBF and

sonicated for an hour. Finally, the Ciprofloxacin® solution was syringed out

and the samples were sonicated with an excess of anhydrous THF and

deionised water (twice for 2 minutes each). The same procedure was followed

to coat Ciprofloxacin® to ODPA SAM coated Ti6Al4V surfaces.

Reaction Mechanism

The reaction scheme in Scheme 4-1 shows the binding of Ciprofloxacin® to

the SAMs through its carboxylic end. Azema et al. [169], Cormier et al. [170]

and Sharma et al. [171] reported the reaction of Ciprofloxacin® with acyl

chlorides. In all these reported procedures, the reaction mainly occurred at the

amine group of the Ciprofloxacin® rather than the carboxylic group to afford

the corresponding amide. This was due to the fact that during all of these

reported reactions a base, triethylamine (R:Et3N), was used. For the synthesis

of amide from acid chlorides via Schotten-Baumann reaction, a base is

essential. This is because the formation of amides from acid chlorides and

amines is accompanied by the production of one equivalent of HCl. This HCl

should be neutralised by a base (such as NaOH and triethylamine0 for the

reaction to proceed further. Thus, by using a base, the reaction can be

controlled to occur through the amine group.

Anions (RCO2-) are better nucleophiles than a neutral nucleophile (-NH-)

[172]. According to the reaction mechanism proposed in this study, the anions

will attack the carbonyl group leading to the formation of Cl- and H

+. These

ions will protonate the amine group of Ciprofloxacin®. As a result of this

protonation, the amine group will acquire a positive charge and become

unreactive. During this attack, the acid will protonate the amine in


Ciprofloxacin® since carboxylate is a poor electrophile and the ammonium ion

is not nucleophilic. Hence the reaction between the acyl chloride and the drug,

Ciprofloxacin® is most likely to occur through the carboxylic group and lead

to the formation of acid anhydride. If the same reaction is considered in the

presence of a base, the acyl chloride will react with the amine, since the HCl

produced during the reaction is neutralised by the base. This will then lead to

the formation of amide rather than acid anhydride.

4.1.3.8. Stability Studies

The stability of the anti-bacterial drug under oxidative exposure and in vitro

conditions was studied for 16-PhDA SAM coated surfaces immobilised with

Ciprofloxacin®.

a) Oxidative Condition

Oxidative stability was determined by exposing the Ciprofloxacin® coated

Ti6Al4V surfaces to ambient laboratory conditions for 7, 14, 28 and 42 days

following a previously reported procedure [58]. After oxidative exposure, the

samples were rinsed in de-ionised water for 1 minute and dried using

compressed air before characterisation. A total of three samples were used to

study the stability at each time interval using an XPS and static water contact

angle goniometry. Survey spectra and high resolution spectra for C 1s, O 1s, N

1s and F 1s were collected using an XPS for all samples and the relative

atomic percentages were averaged. Static contact angles obtained for each

sample were averaged.

b) In vitro Condition

In vitro stability of the drug attached to 16-PhDA SAM coated surfaces was

investigated by immersing each sample in 5 mL of 10 mM Tris-HCl buffer

solution [Tris-HCl solution was prepared by dissolving 1.214 g

Tris(hydroxymethyl)aminomethane in 900 mL of deionised water, adjusting

the pH to 7.4 and making-up the final volume to 1L] and incubated at 37 °C

for 42 days. The immersed samples were removed from the buffer solution at

7, 14, 28 and 42 day time intervals. The surfaces were rinsed for 1 minute with

deionised water and dried using compressed air before characterisation using


XPS and static-contact angle goniometry. Three samples were used for

characterisation from each time interval and the obtained results were

averaged. Similar to the oxidative study, survey and high resolution spectra

were collected for all the samples in each time interval.

4.1.3.9. Drug quantification

It should be noted that the amount of drug coated to the SAM modified

Ti6Al4V surfaces was not calculated based on the difference in the

concentration of the drug (in reaction mixture) before and after

functionalisation. This was mainly because the amount of drug functionalised

to the Ti6Al4V surfaces would be expected in nano-grams. Some amount of

drug may be physisorbed to the surface. Thus, drug quantification by this

method would affect the result. Hence, the total amount of Ciprofloxacin®

functionalised onto SAM coated Ti6Al4V metal surfaces was quantified using

a base hydrolysis method reported previously [37].

The drug coated samples (n=5) were immersed in a 10 mL of 10 mM NaOH

solution (pH ~ 9) and incubated at 37°C. The immersion time interval to

hydrolyse nearly all drug molecules was previously optimised by immersing

the samples for 1 and 2 week time intervals. When characterised using XPS,

after 1 week, a small amount of drug was observed on the Ti6Al4V surface

whereas after 2 weeks of immersion, no drug was observed. Hence the 2 week

time interval was used. The drug eluted solution was adjusted to pH 7.4 using

1 M HCl. A standard stock solution of Ciprofloxacin® (1 mg/mL) was

prepared in 10 mM NaOH buffer solution of pH ~ 7.4. Working solutions

including 1 ng/mL, 10 ng/mL, 50 ng/mL, 100 ng/mL, 250 ng/mL, 500 ng/mL

and 1 µg/mL were prepared with pH ~ 7.4 by diluting the stock solution with

the 10 mM NaOH.

Using a PerkinElmer LAMBDA 35 UV-Visual spectrophotometer (UV-Vis),

the drug dispersed solution was scanned between the wavelengths of 200 to

500 nm. The maximum absorbance value obtained in the UV spectra (λmax)

was 272 nm (Figure 4-5) and is consistent with the previously reported

literature [173–176]. Shoulders were also observed at 217 nm and 325 nm,

close to the theoretical values of 222 nm and 320 nm [177]. The absorbance


peak in the figure 4-5 correlates to the transition of n/π electrons to the π*

excited state. The shoulder towards the higher binding energy is not a double

peak. From the relative absorbance, a calibration curve was plotted for the

working solutions (1 ng/mL – 1 µg/mL of Ciprofloxacin® in 10 mM NaOH of

pH ~ 7.4).

Figure 4-5 Maximum absorbance observed for Ciprofloxacin®

The calibration plot was linear as shown in Figure 4-6 and obeyed Beer-

Lambert’s law. The equation of the line will be Y = Mx + C, where Y is the

absorbance at a given wavelength, M is the slope, x is the concentration and C

is the intercept. Substituting the absorbance value of the unknown

concentration in this equation will give the concentration. By using this

concentration, the total amount of drug was quantified.

Figure 4-6 Calibration Graph for UV-Vis spectrophotometer to determine the unknown

concentration for the drug quantification experiment.

y = 0.000152x + 0.002429 R² = 0.995634

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0 200 400 600 800 1000 1200

Ab

so

rban

ce (

ab

s)

Concentration ng/mL


4.1.3.10. Drug elution studies

Ciprofloxacin® coated Ti6Al4V samples (n=3 for each time interval) were

immersed in 5 mL of 10 mM Tris-HCl buffer solution (pH ~ 7.4) and

incubated at 37 °C for 7, 14, 28 and 42 weeks. Samples were taken out at these

predetermined time intervals. A standard stock solution of Ciprofloxacin (1

mg/mL) was prepared in Tris-HCl buffer solution of pH ~ 7.4. Working

solutions including 1 ng/mL, 10 ng/mL, 50 ng/mL, 100 ng/mL, 250 ng/mL,

500 ng/mL and 1 µg/mL were prepared by diluting the stock solution with the

Tris-HCl buffer.

A calibration curve was plotted (absorbance versus concentration) by

measuring the absorbance values for these standard working solutions (1

ng/mL, 10 ng/mL, 100 ng/mL, 250 ng/mL, 500 ng/mL and 1 µg/mL of

Ciprofloxacin® in 10 mM NaOH of pH ~ 7.4) at the maximum absorbance of

272 nm. The Ciprofloxacin® concentration in the buffer solution was then

calculated by substituting the absorbance of the unknown concentration in the

obtained equation (Figure 4-7).

Figure 4-7 Calibration Graph for UV-Vis spectrophotometer to determine unknown

concentration for the drug elution experiment.

4.1.3.11. Antibacterial susceptibility test

Infections associated with implants are mainly due to microbial contamination.

Some of the most common organisms causing infections are Staphylococcus

y = 0.000126x + 0.001248 R² = 0.998396

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0 200 400 600 800 1000 1200

Ab

so

rban

ce (

ab

s)

Concentration ng/mL


aureus, Staphylococcus epidermidis Pseudomonas aeruginosa and Escherichia

coli [1,178]. Ciprofloxacin®, being a broad spectrum antibiotic, will inhibit the

growth of the majority of the both Gram positive (G+) and Gram negative

bacteria (G-). G+ bacteria are primitive cell organisms that lack a membrane-

bound nucleus. G- bacteria have well-organised cell organelles enclosed within

cell membranes. Hence in this study, the antibacterial susceptibility test was

tested against a G+ bacterium, S. aureus (ATCC 25923) and a G- bacterium, E.

coli (ATCC 25922). The antibacterial susceptibility tests were performed to

qualitatively investigate whether the released Ciprofloxacin® had retained its

active form upon its release and produced a therapeutic effect. It should be

noted that this study does not aim to quantify the drug concentration by this

test or study the minimum inhibitory concentration (MIC) for the bacteria.

a) Test-1

In test-1, Ciprofloxacin® was functionalised to SAM-coated SLM-fabricated

and mechanically polished Ti6Al4V discs of 8 mm diameter using the

procedure for functionalisation mentioned in (4.1.3.7). Among the four discs

used for each plate, two were standard filter paper discs and the other two were

metal discs. Of these four discs one of each metal and the standard disc served

as control with no drug on it. However, they were dipped in ethanol and dried

at 80 °C prior to placing them on the inoculated agar plate. Among the other

two discs, the standard control filter paper disc was coated with approximately

1 ml of 1 mmols (approximately 330 µg/mL) Ciprofloxacin® in ethanol. The

remaining metal disc was coated with Ciprofloxacin® using previously

mentioned procedure in this thesis (4.1.3.7).

Sub-cultures of S. aureus and E. coli were prepared 24 hours prior to

inoculating to the Mueller-Hinton agar plate. Test specimens/discs were placed

on the inoculated plates as shown in Figure 4-8. The plates inoculated with S.

aureus were incubated at 30 °C and the plates inoculated with E. coli were

incubated at 37.5 °C since they are the optimum growth temperature for the

corresponding bacteria [179]. The plates (five for each bacterium) were

examined after 24 hours and zones of inhibitions were measured using a

calliper.


Figure 4-8 Schematic representation of test specimens placed on a culture plate

b) Test-2

Antibacterial sensitivity of the Ciprofloxacin® released to the Tris-HCl buffer

solution was tested against S. aureus (ATCC 25923) and E. coli (ATCC

25922). Drug coated Ti6Al4V cuboid specimens were immersed in 5 ml of

Tris-HCl buffer solution for 7, 14, 28 and 42 days. After these time intervals

the specimens were taken out and the drug eluted buffer solution was used to

determine the sensitivity of the eluted drug. 1 ml of the drug eluted Tris-HCl

buffer solution was syringed out and added drop-by-drop to a standard filter

paper disc. The excess solution was evaporated by drying the specimen in an

oven operating at 80 °C for an hour. A similar procedure was followed to

adsorb the Ciprofloxacin® eluted to the Tris-HCl buffer at different time

intervals to the standard discs. In this study, 3 control discs were used. One

was immersed in ethanol and dried and the other was immersed in Tris-HCl

buffer and dried. The third control disc was coated with 1 ml of 1 mM

(approximately 330 µg/mL) Ciprofloxacin® in ethanol. After drying, the filter

paper disks impregnated with Ciprofloxacin® and the control discs were

placed on the agar plate inoculated with the test bacteria (as shown in Figure 4-

9) to test their susceptibility. Since the study is qualitative, the author did not

aim to quantify the amount of drug present in the discs. The susceptibility test

with these discs was performed using the procedure described in Test-1.


Figure 4-9 Schematic representation of test specimens placed on a culture plate. Label

description: C- control disc with no drug; CD – control with drug on; TC: Control disc coated

with Tris-HCl buffer; 1, 2, 4, 6 – immersion time intervals (in weeks) of the samples in buffer

solution.

4.1.3.12. Statistical Analysis

The experimental results reported in this thesis represent mean ± standard error

of the mean. Standard error was calculated from the results obtained for 5

different samples for contact angle measurement, surface roughness

measurement, particle size analysis and anti-bacterial sensitivity tests, and 3

different samples for the XPS and UV-Vis. The mean and the standard error of

the mean were calculated using Microsoft™ Excel 2013.

Chapter 5: Surface Morphology and Surface Chemistry 79

5. SURFACE MORPHOLOGY AND SURFACE

CHEMISTRY

5.1. Introduction

This chapter concentrates on the surface morphology and surface chemistry of

the Ti6Al4V powders used and the parts that were fabricated by SLM. Surface

morphological and surface chemical effects of recycling Ti6Al4V powders in a

SLM process is discussed. Later, the surface morphology and surface

roughness profile of SLM as-fabricated (SLM-AF) surface is stated.

Comparisons of SLM-AF’s surface chemistry with a SLM fabricated and

mechanically polished (SLM-MP) and a conventionally forged and

mechanically polished (FGD-MP) surfaces are discussed in detail. Finally the

surface chemistries of a laser scanned area, skin scanned and non-skin scanned

areas top and side surfaces, are discussed.

5.2. Effect of Aging of Ti6Al4V Powders

5.2.1. Surface Morphology

Gas atomised Ti6Al4V powders are often used to fabricate samples in SLM

due to their reduced oxide content and their ability to yield high density parts.

In the SLM process, any unused powder material during the process is

recycled to fabricate new parts. Since this exposed powder material is recycled

several times, it is important to study the surface morphology and chemistry of

the powders to investigate if there is any change in their surface properties due

to aging.

The surface morphology of Ti6Al4V powders used to fabricate the samples

were surface characterised before processing in an SLM machine. Figure 5-1

represents the typical surface morphology of virgin and recycled (more than

500 processing hours) powders obtained using an SEM. From the figure it can

be observed that the gas atomised virgin particles had a spherical shape with a

broad size distribution. However, satellites were observed on most of these

spherical particles. The recycled Ti6Al4V powders depicted a small

deformation in this morphology (5-1c and 5-1d). These deformations are due


to the partial melting of particles by the absorption of laser energy. Although,

the observed deformations in the sphericity were not significant, this may

affect the particle size, flow properties of the powder and thus induce

porosities within a part fabricated using these powders. Sieving of the recycled

powder before processing can be beneficial to remove such large lumps of

particles.

Figure 5-1 Surface morphology of virgin (a and b) and recycled (c and d) Ti6Al4V particles

obtained using SEM at different magnifications.

Figure 5-2 shows the morphology of particles retained after sieving the

recycled powders with a mesh size of 63 µm. From the figure it is clear that the

SLM process increases the particle size and affects the particle shape and

sphericity. Agglomeration, partial melting and fusion of particles can be

observed. Although the amount of these retained particles on sieving is low,

sieving is essential to remove these large irregular partially melted particles to

obtain an improved packing density.


Figure 5-2 SEM micrographs of Ti6Al4V recycled powders retained after sieving using a

sieve of diameter of 63 µm.

5.2.2. Particle Size Distribution

The particle size distributions of the virgin, recycled and the sieve-retained

recycled powders were measured using a Malvern Mastersizer to determine if

there was a change on laser processing. It can be observed from the Figure 5-3

that the virgin powders have a slightly narrower size distribution compared to

the recycled powder. Also, the typical size distribution of the retained powder

was significantly higher than the virgin and the recycled powders. As it is

discussed in section 5.2.1, the increase in the size is mainly due to the partial

melting of particles by the absorption of laser energy, presumably on the

fringes of the build area.

Figure 5-3 Comparison of particle size distribution obtained for virgin, recycled and sieve

retained Ti6Al4V particles.

0

2

4

6

8

10

12

14

16

18

10 100 1000

Vo

lum

e D

en

sit

y (

%)

Size Classess (microns)

Virgin Powder

Recycled Powder

Retained Powder


The particle size below which the 10% (Dv 10), 50% (Dv 50) and 90% (Dv

90) of the total volume of the sample exists was calculated to determine the

two points of the distribution that describes the coarsest and finest particles of

the distribution, shown in Table 5-1. Span is the width of the distribution based

on the 10%, 50% and 90% quantile and is calculated using the equation [(Dv

90 - Dv 10) / Dv 50]. These calculations have also shown an increase in PSD

on recycling. The sieve retained sample showed 10% of the particles to be over

140 µm which is predominantly due to the partial melting of particles.

Table 5-1 Calculated particle size distributions obtained for virgin, recycled and sieve retained

powders using a Malvern Mastersizer (average of 5 measurements).

Size Parameter Virgin Powder Recycled Powder Retained Powder

Dv 10 22.4 (µm) 21.4 (µm) 52.6 (µm)

Dv 50 33.2 (µm) 33.4 (µm) 86.1 (µm)

Dv 90 48.7 (µm) 51.8 (µm) 140 (µm)

Mean size, based

on surface area

(D[3,2]) 31.7 (µm) 31.8 (µm) 80.9 (µm)

Mean size, based

on volume (D[4,3]) 34.5 (µm) 60.1 (µm) 117 (µm)

Span 0.79 0.91 1.02

The results show that on recycling the median size of the particles has

increased due to the partial melting of particles. This partial melting could be

due to several factors including the absorption of thermal energy by the

surrounding particles in the build area. Another possible reason could be in

cases where SLM failed to produce parts due to the lack of support structures.

For example, if there are no supports to hold the particles in position, then, the

laser beam will melt particles in the build area but will not sinter this to a

previously scanned layer/solid structure. Hence, particles will float and will

lead to the formation of lumps of particles.

Partial melting was observed to affect the sphericity and lead to irregularly-

shaped particles. Although this change in size distribution and morphology

was not significant, there is a chance that these particles, if present on the build

area, may affect the part’s density by inducing pores. A particle size

distribution of 20 µm – 50 µm has been observed to render a high flowability


and highly packed powder bed density since the smaller particles filled in the

gaps between the relatively large particle sizes [81]. A similar increase in the

PSDs of Ti6Al4V processed in a SLM machine was also witnessed by Seyda

et al. [83]. Hence sieving of particles before processing is required.

5.2.3. Surface Chemistry

The surface chemistry of virgin and recycled Ti6Al4V powders was studied

using XPS to investigate the effect of aging on the surface chemistry of the

Ti6Al4V powders. Figure 5-4 shows the survey spectra of virgin and recycled

Ti6Al4V powders obtained using XPS. The spectra show the presence of

major alloying elements titanium and aluminium. Vanadium (V) was not

observed in the virgin powder whereas a very small concentration was

observed on the surface of recycled powders. The relative atomic percentages

obtained for virgin and recycled powders obtained using XPS were tabulated

in Table 5-2. Figure 5-5 shows the high resolution XPS spectra for C 1s, O 1s

and N 1s regions for both virgin and recycled powders.

Figure 5-4 Survey Spectra obtained for virgin and recycled Ti6Al4V powders using XPS.


Table 5-2 Relative atomic percentages obtained for virgin and recycled powders using XPS.

The calibration error during peak integration is ± 10%.

Metal

Relative atomic percentage

Virgin Recycled

Al 2p 9.41 ± 0.2 9.16 ± 0.6

C 1s 18.95 ± 0.9 23.75 ± 1

V 2p 0 0.31 ± 0.21

O 1s 55.12 ± 0.6 50.5 ± 0.1

Ti 2p 15.13 ± 0.5 14.21 ± 1

N 1s 1.15 ± 0.2 2.06 ± 0.2

Figure 5-5 High resolution spectra of C 1s, O 1s and N 1s regions obtained for virgin and

recycled Ti6Al4V powders using XPS.


It can be observed from the figure that for both virgin and recycled Ti6Al4V

powders, the deconvoluted C 1s spectra showed the possibility for peaks at 285

± 0.2 eV, 286.6 ± 0.1 eV and 289.2 ± 0.2 eV. These peaks correspond to C-C,

C-O and O-C=O which are mainly due to contamination of the powders by

hydrocarbons [144,180]. For O 1s, the peaks observed at 530.2 ± 0.1 eV, 530.8

± 0.1 eV and at 532 ± 0.3 eV can be attributed to metal oxide (titanium oxide),

Al-O and/or C-O and O-H and/or C=O respectively [155]. The N 1s peak at

396.6 ± 0.1 eV is due to metal nitride and 400.1 ± 0.1 eV can be attributed to

C-N surface contaminants [181]. An increase in the C and N (marked with

dotted lines in the Figure 5-6) observed for recycled Ti6Al4V powders are

mainly due to contamination.

Figure 5-6 Comparison of high resolution spectra of C 1s, O 1s and N 1s regions obtained for

virgin and recycled Ti6Al4V powders using XPS. The dotted lines show a possible change in

the elemental composition of recycled powders due to contamination from the atmosphere.

Figure 5-7 shows the deconvoluted high resolution XPS spectra for Ti 2p, Al

2p and V 2p regions of both virgin and recycled powders. Deconvoluted Ti 2p

peak showed the presence of titanium in both their metallic and oxidation

states. A significant contribution from TiO2 was observed due to the

characteristic doublet Ti4+

peak at 458.8 ± 0.1 eV and 464.3 ± 0.1 eV for both

virgin and the recycled powder. Ti2+

(455.7 ± 0.1 and 460.2 ± 0.1 eV) and the

underlying Timetallic

at 454 ± 0.1 eV and 460 ± 0.1 eV were also witnessed.

According to previous literature, the presence of Ti3+

would be expected on

both Ti6Al4V powders, whereas the deconvoluted peaks did not show their

existence [155]. This may be due to differences in the peak deconvolution.


Figure 5-7 High resolution spectra of Ti 2p, Al 2p and V 2p regions obtained for virgin and

recycled Ti6Al4V powders using XPS.

Aluminium observed on the virgin powder was in its sapphire form of Al2O3

(74.4 eV) whereas the recycled powder showed Al to be both in its metallic

(71.9 eV) and sapphire form (74.5 eV). The survey spectrum of virgin powder

did not show the existence of vanadium whereas the high resolution V 2p peak

showed the possibility for the occurrence of vanadium in its oxide form on the

virgin power. Due to the lower intensity of the peak compared to the recycled

powders, the survey spectrum did not exhibit the presence of vanadium.

However, the concentration of V on the recycled powder was higher than for

the virgin powder and hence the survey spectrum predicted the existence of V

on the recycled powder. The observed V on both powders was predominantly

from its oxide; however, an inference for the existence of V in its metallic form

(at 512. 1 eV) was observed.


The surface characterisation using XPS showed no significant change in the

chemistry of the virgin and recycled powders apart from contamination. The

increase in the C and N observed for recycled Ti6Al4V powders were mainly

due to contamination [132]. The formation of titanium nitride and TixOyNz

(titanium-oxy-nitride) may be due to the contamination in the gas/atmosphere

to which the powders were exposed to. Contamination of powders is possible

due to atmospheric contaminants and other possible contaminants such as oils

(from pump/motor) from the SLM machine and manual handling [146].

Previous studies suggest that entrapment of carbon monoxide and carbon

dioxide due to the oxidation of carbon was observed to induce porosity [91].

Also, titanium is highly reactive and can form titanium carbide [182]. This can

affect the wettability of the fabricated layer and in turn affect the mechanical

properties of the part manufactured by SLM. Since parts are produced

additively in the SLM process, wettability of the surface will have a significant

impact on the density of a part [92].

5.3. SLM Surface Characterisation

5.3.1. Surface Morphology and Roughness

Figure 5-8 shows the surface morphology of a SLM-AF surface at various

magnifications. From the micrograph it is can be noted that loosely bound and

partially melted particles are present on the surface. These partially melted

particles will add to the thickness of the part and thus increasing its wall

thickness and surface area. The presence of partially melted particles on the

part surface is inevitable in SLM, with the magnitude of this problem being

dependent on both material and process parameters. Some of the parameters

that influence the partial melted /surface roughness of SLM fabricated parts are

material, powder particle size, layer thickness, laser type and power, scan

parameters and scan strategy [61,95].


Figure 5-8 SEM images showing the surface morphology of an as-fabricated Ti6Al4V surface

fabricated using SLM. The figure clearly depicts the rough and porous nature of the SLM-AF

surface.

The observed surface roughness pattern and the surface roughness (Ra)

obtained using a surface profilometer was 17.6 ± 1.66 µm. However, it was

observed to be difficult to characterise the roughness in certain areas of the part

due to the presence of large voids as shown in Figure 5-9. The white voids in

the figure represent the areas from which signals were not obtained during

surface profiling.

Figure 5-9 Surface Topography of an SLM as-fabricated Ti6Al4V surface obtained using a

surface profilometer. The measured surface area was 175 µm2.

The microstructure of the SLM fabricated part was not studied since it was out

of this project scope. However, previous literature reported that SLM


fabricated components consist of α martensitic phase precipitated from β

columnar grains [88].

5.3.2. Surface Chemistry

5.3.2.1. Surface film

The surface chemistry of SLM as-fabricated (SLM-AF), SLM fabricated and

mechanically polished (SLM-MP) Ti6Al4V components were studied and

compared with a forged and mechanically polished (FGD-MP) Ti6Al4V

component. The surface chemistry of a vertical planar face (side face) of the

cuboid samples was studied using XPS.

Initially, survey spectra for the SLM-AF, SLM-MP and FGD-MP samples

were obtained to determine their surface chemistry. Figure 5-10 shows the

survey spectra of the detected elements for all three samples. It is evident from

the spectra that the elements C, N, O, Ti and Al are present in all three

samples. A weak signal (≥ 0.3%) for vanadium was observed on the

mechanically polished samples (SLM-MP and FGD-MP) and not on the SLM-

AF. Since the SLM-AF samples were rough, this would have reduced the

signal by reducing the electron counts reaching the detector during XPS

characterisation and hence lead to lack of evidence for no vanadium on the

SLM-AF surface.

Figure 5-10 Survey spectra of SLM as-fabricated (SLM-AF), SLM mechanically polished

(SLM-MP) and forged mechanically polished (FGD-MP) Ti6Al4V surface.


Table 5-3 shows the relative atomic percentages of all detected elements. As

can be observed from the table, the dominant signals of the spectra for all three

samples are Ti, O and C with a weak contribution from Al and N. The ratio of

Ti:O observed for SLM-AF (0.32 ± 0.03), SLM-MP (0.27 ± 0.03) and FGD-

MP (0.26 ± 0.02) were higher than Al:O (0.11 ± 0.01, 0.1 ± 0.01 and 0.06 ±

0.01 for SLM-AF, SLM-MP and FGD-MP) ratio. As the Ti:O ratio for all three

samples was higher than its corresponding Al:O ratios, this revealed that the

surface film of all three samples mainly composed of titanium oxide. To

investigate the surface chemistry in detail and to determine the oxidation states

of alloying elements, high resolution spectra of C, O, Al, Ti and V on the

SLM-AF, SLM-MP and FGD-MP samples were obtained. Depth profiling was

performed to investigate the distribution of the detected elements and their

chemical states in the first few nanometres of the Ti6Al4V components.

Table 5-3 Relative atomic percentage of elements detected using XPS for SLM-AF, SLM-MP

and FGD-MP samples. The calibration error during peak integration is ± 10%.

Sample

Type

Relative Atomic Percentage

Al 2p C 1s N 1s O 1s Ti 2p V 2p

SLM-AF 5.4 ± 0.5 25.8 ± 2.1 3.2 ± 0.8 49.8 ± 1.6 15.8 ± 0.9 0

SLM-MP 6.0 ± 0.6 17.2 ± 2.7 1.2 ± 0.5 59.1 ± 2.1 16.1 ± 0.5 0.4 ± 0.1

FGD-MP 3.0 ± 0.9 32.6 ± 2.3 1.8 ± 0.3 49.2 ± 1.4 13.2 ± 1.2 0.3 ± 0.2

5.3.2.2. Depth Profiling

Figure 5-11 shows the high resolution and depth profiling spectra of C and O

for all three SLM-AF, SLM-MP and FGD-MP samples. Bulk metals react with

air and form their corresponding oxides, carbides and nitrides depending on the

atmospheric conditions [183]. Although the SLM process is carried out in an

argon atmosphere with oxygen concentration < 1 ppm, there is the possibility

of titanium reacting with the available oxygen and forming its oxide. Also,

there is less control of carbon dioxide/nitrogen during the SLM process and the

actual amount of these elements present in the build atmosphere/chamber

remains unknown. Although Ti6Al4V has 0.08% (max) carbon in its actual

composition, a considerable amount of carbon is present in the powder due to

atmospheric contamination. Characterising a surface in XPS without carbon

contamination is almost impossible.


Figure 5-11 Depth profile spectra of C 1s and O 1s regions for SLM-AF, SLM-MP and FGD-

MP samples. Depth profiling was performed every 30 seconds (~ 4 nm) from 0 to 270 seconds.

C 1s spectra show the disappearance of the initially observed peak after 30 seconds of etching

which is attributed to surface contamination. Reduction in the intensity of O 1s peak on

increasing depth can also be witnessed from the figure.

Before etching the sample, the C 1s peak showed the possibility for the

existence of metal carbide (281.7 ± 0.2 eV), C-C (285 ± 0.1 eV) and O=C-O

(288.9 ± 0.3 eV) bonds for all three samples [155]. Since Ti6Al4V is an alloy

of Ti, Al and V, the observed metal carbide may have contributions from all

three metals. However, since Ti is highly reactive and is present on the surface

at a higher concentration, most of the observed metal carbide may have been

due to titanium carbide. After etching the samples for 30 seconds (~ 4 nm),


SLM-AF showed a reduction of C-C and C-O peaks whereas in SLM-MP and

FGD-MP samples, the C-C and C-O peaks disappeared. The XPS detected C-

C, O-C=O bonds can be due to hydrocarbons from air, manual handling,

solvents used for cleaning purposes [183,184]. Since solvents including

dichloromethane, methanol and ethanol were used for cleaning purposes, these

solvents may have left behind some carbon residues that were not washed

away by deionised water. Also, there is the possibility for contamination due to

adventitious carbon in the atmosphere while mechanical polishing, mounting

and transferring the sample into the XPS machine [181]. On sputtering the

surface, the concentration of carbon decreased significantly in the first few

nanometres. This clearly suggests that they are mainly from contamination. A

slower decrease in the C for SLM-AF might be due to the rough nature of the

SLM-AF surface. On depth profiling the sample, only the metal carbide peak

was observed. Thus, depth profiling of the sample provided a powerful way to

separate out the carbon contribution from atmospheric contamination.

The metal carbide peak (at 281.7 ± 0.2 eV) was witnessed even after etching

all three samples to several nanometres. This shows the possibility for their

formation during a laser scan and is not purely from contamination. Due to the

higher reactivity of Ti, it may readily react with carbon in the build chamber to

form its carbide. An exothermic formation of oxides and carbides of Ti during

SLM process was previously reported to play a significant role in the

instability of the melt pool [91]. As a result, the formation of irregular pores

was witnessed.

The survey spectra (Figure 5-10) showed the presence of N on all three

samples in addition to C. Previous studies have shown that the contribution to

the nitrogen peak can be from organically and inorganically bound N. the

presence nitrogen in the feed powder has been discussed in section 5.2.

Inorganically bound N may originate from a small amount of TiNx or TixOxNx

in the surface film since titanium readily reacts with nitrogen. Organically

bound nitrogen could be from organic contaminants such as proteins from

manual handling or the atmosphere [183]. However, similar to C, the observed

metal nitride may also have contributions from Al and V in addition to Ti.


On examining the high resolution O 1s spectra, the possibility of peaks at

530.2 ± 0.1 eV and 532.3 ± 0.2 eV were observed (Figure 5-11). The initial

peak is due to Ti-O and the peak observed at 532.3 ± 0.2 eV may be due to

C=O, aluminium oxide and adsorbed water molecules [155]. The presence of

adsorbed H2O or hydroxyl molecules could be due to the final cleaning

procedure (using deionised water) or from atmospheric moisture. The Ti-O

bond is due to the thin TiO film formed on the surface and C=O may be due to

the contamination of the surface with carbon containing molecules [183]. From

the depth profiles, the concentration of Ti-O can be observed to decrease

rapidly with depth for the SLM-MP and FGD-MP samples while it reduces

slowly for the SLM-AF. The possible reasons for this sudden and slow drop of

oxygen concentration are due to the effect of surface roughness pattern and are

discussed in section 5.3.2.3.

Figure 5-12 shows the high resolution Ti 2p peak for SLM-AF and Figure 5-13

shows its evolution on depth profiling. SLM-MP and FGD-MP also showed a

similar high resolution Ti 2p peak to SLM-AF. A characteristic doublet peak

shape and position observed at 459.1 ± 0.2 and 465.5 ± 0.2 eV for all three

samples showed that Ti is mainly present on the surface as Ti4+

[183]. Hence,

it is evident that the surface of the samples is dominated by TiO2. Minor

fractions of its sub-oxides Ti2O3 (457.8 ± 0.2 and 464.2 ± 0.1) and TiO (455.1

± 0.2 eV and 460.9 ± 0.1 eV) were observed from the intensity of Ti 2P peak

revealing their presence in the metal-oxide interface [155]. A small

contribution from the Ti 2p metal (453.8 ± 0.1 eV and 460.1 ± 0.2 eV) was

witnessed and this is due to the fact that the oxide is thinner than the electron

escape depth [183]. On depth profiling of the samples, a clear shift of the TiO2

peak to pure Ti metal doublet peak at 454.5 ± 0.1 eV and 460.6 ± 0.2 eV could

be witnessed (Figure 5-13). This confirms that TiO was only present in the first

few nanometres.


Figure 5-12 High resolution XPS scan of Ti 2p region showing the oxidation states of titanium

on a Ti6Al4V SLM-AF surface.

Figure 5-13 Depth profiles obtained for titanium 2p region using XPS for SLM-AF, SLM-MP

and FGD-MP surfaces. Depth profiling was performed at every 30 seconds (approximately 4

nm) from 0 to 270 seconds.

Ti has a high affinity towards O and moreover O has high solubility in Ti

[185]. When Ti is exposed to O, it forms a monolayer of oxygen on the metal

surface. The oxide is then formed by the diffusion of O atoms into the metal

and/or metal ion diffusion on to the surface. After nucleation, oxide grows

laterally to cover the whole surface [185]. This reaction continues until the

diffusion of O ceases leading to the formation of a passive oxide film. Since

SLM-AF surface is porous (17.6 ± 1.66 µm), it is possible that there is a

greater chance for more oxygen to diffuse into the metal and increase the oxide

layer thickness. A previous study revealed that recycling of Ti6Al4V powders


in an additive manufacturing machine increased the thickness of its surface

oxide layer from 96 Å to 112 Å [163].

The spectra obtained for Al in the Al2P region shows the possibility for the

presence of Al both in its metallic and oxide form for all three samples (Figure

5-14). Peaks observed at 71.8 ± 0.1 revealed the presence of metallic

aluminium from the underlying bulk metal alloy and 74.6 ± 0.2 showed the

presence of Al2P3 in the sapphire form of Al3+

(Al2O3) [155]. Also this Al3+

ion

may exist as interstitial or substitutional ions in the TiO2 matrix [155]. Similar

to the Ti 2p peak, on depth profiling, an increase in the metallic Al and a

decrease in its oxide form was observed for all three samples. However, the

SLM-MP and FGD-MP showed a sudden shift from the oxide form to metallic

Al whereas SLM-AF witnessed a slower change. The transformation of Al3+

ion to metallic Al on depth profiling revealed the presence of aluminium oxide

only in the first few atomic layers.

Figure 5-14 Depth profiles obtained for aluminium 2p region using XPS for SLM-AF, SLM-

MP and FGD-MP surfaces. Depth profiling was performed at every 30 seconds (approximately

4 nm) from 0 to 270 seconds.

From Figure 5-15, it can be observed that there is no V on the SLM-AF surface

whereas it is observed on the SLM-MP and FGD-MP surfaces in low

concentration. This could be due to the fact that the SLM-AF surface has

comparatively higher Ti and Al concentration than the V concentration. With

this very low concentration, the signals generated would be very low and this

might have also been affected by the rough nature of the surface. Although V

was not observed on the SLM-AF surface, after mechanical polishing, the


alloying element is exposed to the atmosphere form its corresponding oxides

(V2O3 and V2O5).

Figure 5-15 Depth profiles obtained for vanadium 2p region using XPS for SLM-AF, SLM-

MP and FGD-MP surfaces. Depth profiling was performed at every 30 seconds (approximately

4 nm) from 0 to 270 seconds.

The concentration of V on the SLM-MP was lower than the concentration of V

on FGD-MP. However, most of the V observed on the surface was recorded to

be in its oxide form (V2O3 and V2O5) at 514.7 ± 0.2 eV and 516.8 ± 0.3 eV for

both SLM-MP and FGD-MP [183]. On depth profiling, XPS detected V in the

SLM-AF sample (after sputtering ~ 4 nm) and the concentration of metallic

vanadium (512 ± 0.1 eV) could be observed to increase for all three samples as

a function of sputtering time.

5.3.2.3. Estimation of oxide layer thickness

All three surfaces were observed to have TiO2 as the dominant surface film.

Due to the presence of Al3+

ions on all three surfaces, it is likely that the

surfaces also contain Al2O3 [155,183,185]. However, the implied concentration

of Al2O3 on all these surfaces was lower than the implied concentration of

TiO2. The presence of Al2O3 on the surface film will add to the corrosion

resistance of Ti6Al4V. Vanadium was not observed on the SLM-AF, whereas

SLM-MP and FGD-MP surfaces were observed to have vanadium on their

surface as V2O3 and V2O5.

Figure 5-16 shows the graphical representation of the elemental distribution

observed for SLM-AF, SLM-MP and FGD-MP samples on depth profiling.

Since O concentration reduced slowly for SLM-AF due to the rough nature of


the surface, the actual thickness of its oxide layer is difficult to calculate by

both of these methods. Also, it should be noted that the methods used for

estimating the oxide layer thickness do not produce directly comparable

results; however, comparison may be made between the samples using the

results obtained by one method.

Figure 5-16 Estimation of oxide later thickness for SLM-AF, SLM-MP and FGD-MP samples

using two different methods. Data points at every 40 angstroms.

The rough nature of the SLM-AF surface is due to the partial melting of

particles. On mechanical polishing of the as-fabricated part, these partially

melted particles can be removed from the surface. The rough and porous nature

of the SLM-AF is likely to affect the XPS measurement. An important

parameter to consider while estimating the oxide film thickness would be

sensitivity of XPS to rough surfaces. During the measurement, the number of


counts obtained for a rough surface (SLM-AF) was considerably lower than

the counts obtained for a polished surface (SLM-MP and FGD-MP). Since the

surface is highly porous and uneven due to partially melted particles, this gives

varied incident angles for the XPS beam (see Figure 5-17). Hence, the number

of photoelectrons that pass through the analyser will be significantly lower

when compared to a polished surface. When the surface morphology is uneven,

it is difficult to sputter the surface equally due to a shadowing effect. As a

result, some elements present on the outermost surface can still be detected

even in a deeper location in the depth profile. Due to this shadowing effect and

uneven etching of the sample surface, the SLM-AF samples showed a gradual

change in its detected elemental composition.

Figure 5-17 Schematic of XPS probing on a rough SLM-AF Ti6Al4V component.

The values obtained for SLM-AF using method 1 shows it has a thick oxide

layer (> 35 nm). In the SLM process, parts are built layer-by-layer and this

means that every layer will be exposed to the atmosphere within the build

chamber for few seconds until a fresh layer is laid. Although argon gas is used

to inert the SLM process chamber, there is still the possibility for the presence

of oxygen at a very small concentration (less than 8 ppm). This oxygen may

react with the laser scanned area to saturate the layer whereas this may not be

the case in forging. Hence the process conditions might be a reason for the

formation of thick oxide on the SLM fabricated samples. However, the gradual

change in the elemental composition of Ti and O compared to the

mechanically polished surfaces (SLM-MP and SLM-AF) during depth


profiling shows that XPS characterisation of the sample was affected by the

rough nature of the SLM-AF. Thus, the oxide layer thickness of SLM-AF was

observed to be higher than mechanically polished surfaces.

On comparing the oxide layer thickness for the mechanically polished samples

using method 1, FGD-MP showed thicker oxide layer (~ 15 nm) than SLM-MP

(~ 10 nm). Although this may be true, one of the reasons for this might have

been the effect of high carbon contamination observed on the FGD-MP

surface. The carbon contaminant present above the surface oxide layer might

have masked certain concentration of the oxygen initially. This masking effect

can be clearly observed from the depth profiling of FGD-MP (an increase in

the oxygen concentration as carbon concentration drops). Thus depth

profiling/etching of a metal sample’s surface renders useful information to

examine if the carbon present on the surface is due to contamination or from its

alloy.

In method 2, SLM-MP (~ 6.5 nm) and FGD-MP (~ 7 nm) showed nearly

similar thickness value and lesser than SLM-AF (~ 15 nm). Since these SLM-

MP and FGD-MP samples were mechanically polished, the oxide layer formed

may not be representative of the true oxide layer formed during the

manufacturing process. Also, due to mechanical polishing of the samples, there

will be no or very minimal shadowing effect caused due to partially melted

particles during etching and hence their values are closer. The oxide layer

thickness observed for SLM-MP and FGD-MP were nearly consistent with

literature showing an oxide layer of 1 – 10 nm [186] for un-treated Ti6Al4V

substrates. Thus the surface composition was in the order of metal oxides

leading to bulk metal.

The presence of surface oxide layer on both the SLM-AF and SLM-MP

Ti6Al4V surfaces will be the key for the corrosion resistance of the SLM

fabricated surfaces. The mechanically polished surfaces show a similar oxide

thickness as the conventionally fabricated parts (without any treatment) [186].

However, the SLM-AF shows a thick oxide layer and this could be beneficial

for various applications. If required, to improve the surface oxide thickness,

further surface modifications can be performed. Surface treatments such as


chemical etching, passivation, ion implantation and thermal oxidation are

performed to enhance this oxide layer thickness for biomedical applications

[4,183].

Variola et al. [155] etched Ti6Al4V surface with a mixture of sulphuric acid

and hydrogen peroxide to enhance the oxide layer thickness from 10 nm to 45

nm. Their study reported an enhanced growth of osteoblast cells on the

Ti6Al4V samples with a thick surface oxide film. Anodic oxidation of

Ti6Al4V ELI using an electrolyte solution of 0.5 M H3PO4 was reported to

yield oxide layers in the range of 30 – 120 nm depending on the anodising

current density [187]. Velten et al. [188] performed a thermal oxidation of

Ti6Al4V substrates between 400 and 600 °C for 5 – 500 minutes and reported

the formation of 190 nm thick oxide layer. They have also reported the

formation of oxide films of more than 100 nm thick on the Ti6Al4V surfaces

by anodic oxidation and sol-gel process. Although the SLM-MP parts had an

oxide thickness of less than 10 nm, if a thickness of > 10 nm is required, then

these can be improved by adopting a suitable chemical treatment method.

However, the SLM-AF surfaces show an oxide layer thickness of 15 nm which

is better than the oxide-layer thickness of the SLM-MP and FGD-MP surfaces.

5.4. Surface Chemistry Comparisons

5.4.1. Effect of skin scanning

Surface Morphology

Skin scanning or selective remelting of the top layer of a SLM fabricated

component is performed to increase the surface quality [93]. During skin

scanning, the top/final layer is scanned twice. Figure 5-18 shows the surface

morphology of a non-skin scanned and skin scanned part fabricated using

SLM. Partial melting of particles to the build during SLM (side surface) can be

observed. The non-skin scanned part had more particles sintered to the laser

scanned surface than the skin scanned part. The surface morphology also

clearly shows the laser scan track during the SLM process. An elevated ridge

of the solidified material on the edges of both the NSK and SK parts can be

observed from Figure 5-19. This is mainly because the remelted material is


partially pushed by the laser beam to the contour of the part. A similar effect

has been reported by Kruth et al. [51] for SLM produced parts.

Figure 5-18 Surface morphology of non-skin scanned and skin scanned Ti6Al4V SLM

surfaces depicting the laser scan tracts.

Figure 5-19 Surface morphology of a non-skin scanned (NSK) (a) and skin scanned (SK) (b)

Ti6Al4V part fabricated by SLM.

Skin scanning or selective re-melting of layers during SLM is performed to

increase the surface quality and density of the fabricated part [93]. This is

achieved by remelting the laser-scanned layer to form an even surface. When

the surface is even, the distribution of powder to build the next layer will be

more homogenous and hence will reduce the entrapment of air. Hence, highly

dense parts can be produced. In this study, both the NSK and SK parts

fabricated using the AM 250 were observed to be rough (Ra = 3.4 µm ± 0.24

µm for NSK and Ra = 2.2 µm ± 0.16 µm for SK) due to the presence of

partially melted particles and the laser scan pattern.

Apart from the partially sintered particles, laser scan tracks are clearly visible

in the surface morphology of the NSK and especially in the SK surfaces

(Figure 5-18). Since the SK surfaces are scanned by the laser twice (using a

meander scan strategy i.e. rotating the laser scan to 67°), a relatively smooth


(Ra = 2.2 µm ± 0.2) surface compared to the NSK (Ra = 3.4 µm ± 0.2 µm)

surface was obtained. However, it should also be noted that this re-melting

leaves a laser scan track that might potentially affect the surface profile.

Although optimising the hatch distance may potentially reduce the laser scan

track, this is almost difficult to overcome.

Surface Chemistry

Figures 5-20 shows the high resolution C 1s, N 1s and O 1s spectral regions

revealing the evolution of these constituent elements over the first few atomic

layers (56 nm) of the NSK and SK components. On deconvoluting the C 1s

region (Figure 3) of NSK and SK surfaces, peaks were observed at 285 ± 0.2

eV, 286.7 ± 0.1 eV and 288.2 ± 0.2 eV. These peaks were attributed to C-C, C-

O/C-N and C=O [155]. For both NSK and SK surfaces, carbon observed in

their outermost layer vanished on etching the surface for ~ 4 nm. As mentioned

in the previous section, the disappearance of the C-C, O/C-N and C=O peaks

after etching confirms that the carbon is mainly from contamination of the

surface [132]. However a peak at 282 eV was observed for both NSK and SK

sample after etching their outermost layer. This peak was attributed to metal

carbide [183].

The N 1s region on deconvolution exhibited the possibility for peaks at 396.5 ±

0.2 eV and 400.2 ± 0.2 eV which corresponds to inorganically (metal-nitrogen)

and organically (C-N) bound nitrogen to the surface [181]. On etching the

surface for approximately 4 nm, the organically bound nitrogen disappeared.

This reveals that this organic nitrogen is from nitrogen containing carbon

contaminants (oil from pumps). The deconvoluted O 1s peak exhibited the

possibility for presence of TiOx, C-O and adsorbed water molecules for both

NSK and SK samples. From the Figure 5-20, in addition to the formation of

metal oxides, on depth profiling, the formation of metal carbide and metal

nitride on both NSK and SK surfaces was observed. Although argon is used to

make the chamber inert, there is the possibility for the presence of traces of C

and N in addition to oxygen. AM 250 has a built-in sensor to quantify the

amount of oxygen in the chamber; but not for C and N. Also, C and N are

present in the atmosphere as organic contaminants and as gases (CO/CO2 and


N2). These elements were observed on the Ti6Al4V powders as contaminants

(discussed in section 5.2). Due to the high reactivity of Ti and Al towards

carbon and nitrogen these metal carbides and nitrides are formed [183].

Figure 5-20 XPS spectra of C 1s, N 1s and O 1s regions for the non-skin scanned (NSK) and

skin scanned (SK) Ti6Al4V surfaces fabricated by SLM. The dotted lines shows the peaks that

disappeared on etching the surface.

Figures 5-21 shows the high resolution Ti 2p, Al 2p and V 2p spectral regions

on depth profiling the NSK and SK components. On depth profiling, a clear

transition of the metal oxide to pure metal was observed (circled with dotted

lines in Figure 5-21) for both NSK and SK surfaces. However, the


transformation of Al oxide to aluminium was more gradual for SK compared

to NSK (circled in Figure 5-21 with dotted lines). This slow transition may be

due to the formation of thick oxide of aluminium.

Figure 5-21 XPS spectra of Ti 2p, Al 2p and V 2p regions for the non-skin scanned (NSK) and

skin scanned (SK) Ti6Al4V surfaces fabricated by SLM. The dotted lines show the transition

of metal oxides from the outermost layer to pure metals.

Figure 5-22 shows the evolution of elements on depth profiling of NSK and

SK Ti6Al4V samples using XPS. From the figure it can be observed that on

skin scanning, the concentration of aluminium on the surface increases (nearly

doubled). The thickness of the oxide layer was measured by two different


methods. As mentioned previously, in the first method, the thickness was

measured based on the point at which the concentration of the major alloying

element Ti was equal to that of oxygen. In the other method, the thickness was

taken as the point at which the maximum observed concentration of oxygen

has reduced to half. Using the initial method, the NSK showed a thickness of

approximately 7 nm whereas the SK part showed approximately 14 nm which

is nearly double that of the NSK’s oxide layer thickness. In method 2, for the

NSK surface, the initially observed maximum O concentration of 46% reduced

to 23% at nearly 35 nm; whereas for the SK surface, the initial maximum O

concentration of nearly 47% did not reduce to half until the sampling depth of

56 nm. Thus the oxide layer thickness and the evolution of elements including

Ti, C, N and O on the NSK and SK surfaces were observed to be significantly

different.

Previous literature has suggested the use of skin scanning (re-melting) for

improved surface quality and mechanical properties. However, the surface

chemistry of such skin scanned surfaces has not been discussed [91]. During

skin scanning or re-melting, the top/final layer is scanned twice. Hence during

this scanning period, a high amount of energy will be transferred to re-melt and

solidify the upper most surface. On re-melting, due to the increase in

temperature, segregation of elements in an alloy is possible. For example,

segregation of aluminium was reported due to rapid cooling/solidification [51].

This study observed a high concentration of aluminium in the SK surface

compared to the NSK surface. Although the Ti6Al4V alloy has only 6% Al,

more than 10% on the surface and over 15% into the SK component was

observed; whereas for the NSK component, Al was relatively less abundant

near the surface. Since the solubility of Al in Ti is very low, precipitation of a

Ti3Al phase is possible as the temperature reaches 500-600 ̊C. On re-melting

the previously scanned layer, the material will remain at a high temperature for

a long time leading to precipitation [51]. However, the XPS determined

surface concentration did not show an increase in the metallic titanium

concentration.


Both NSK and SK surfaces mainly consisted of oxides Ti and Al; however,

their concentrations were significantly different when compared to each other.

Ti6Al4V alloy is preferred for various biomedical applications because of the

high cytocompatibility offered by the presence of corrosion resistant TiO2 film.

Also, SLM is considered to be one of the viable processes for making

customised metallic implants with complex structures. The application of SLM

to fabricate implants will be limited if the process affects the surface chemical

composition. Small changes in the chemical composition may cause

catastrophic loss of ductility, corrosion resistance, toxicity and

cytocompatibility [3,51].

Figure 5-22 Evolution of the surface chemistry of non-skin scanned (a) and skin scanned (b)

SLM fabricated surface.


Thus selective remelting of the final layer may be advantageous in terms of

rendering a better surface finish. However, it should be noted that it may also

alter the surface chemical composition and surface oxide layer. The presence

of TiO2 as the predominant surface oxide layer on Ti6Al4V surface is

responsible for its corrosion resistance and cytocompatibility. Since selective

remelting alters this composition by increasing aluminium oxide concentration

on the surface, cytocompatibility of these components may be reduced and

should be properly investigated.

5.4.2. Top Surface and Side Surface

The surface morphology of a side surface and a top surface are shown in

Figure 5-23. It can be observed from the figure that the side surface is very

rough and porous compared to the top surface. As described earlier in section

5.3.1, this is mainly due to the partial melting of particles from the powder bed

to the build. However, the top surface was also accounted for a small number

of partially melted particles. These particles might have been blown away by

the gassing unit after laser scanning on these areas. In addition to the sintered

particles, peaks and valleys due to laser scanning and laser scan tracks can also

be seen in the Figure 5-23b.

Figure 5-23 SEM micrographs of a side surface (a) and top surface (b).

The surface chemistry of the side surface and the top surface was examined

using XPS to determine if there had been any change in the surface chemistry

(Table 5-4). Both surfaces showed high carbon and nitrogen contamination.

The side surface showed the titanium oxide as its major surface oxide film

with a minor contribution from aluminium oxide (Ti:O = 0.32 ± 0.02; Al:O =


0.11 ± 0.01). On the other hand, the oxide film on the top surface (laser

scanned area) was observed to have nearly the same contribution from both

titanium oxides and aluminium oxides (Ti:O = 0.26 ± 0.03; Al:O = 0.24 ±

0.02). On comparing the oxide films of top and side surfaces, it was clear that

the aluminium oxide concentration significantly increased for the top surface.

The reason for more aluminium on the top surface may be due to the

segregation of the metal during laser scanning [51]. The existence of vanadium

on the top surface was observed but not on the side surface. Thus the surface

morphology and surface chemistry of the top surface and the side surfaces

were significantly different. However, it should be noted that the high surface

roughness of the side surface might have affected the XPS results in

determining the exact surface chemistry. Further characterisation of these

surfaces using SEM with energy dispersive X-ray, X-ray diffraction and

transmission electron microscopy is essential to understand the surface

properties in detail.

Table 5-4 Relative atomic percentage of elemental contributions obtained using an XPS. The

calibration error during peak integration is ± 10%.

Sample

Type



Side

Surface 5.4 ± 0.5 25.8 ± 2.1 3.2 ± 0.6 49.8 ± 1.6 15.8 ± 0.9 0

Top

Surface 10.4 ± 0.9 27.9 ± 1.5 6.0 ± 0.7 44.2 ± 1.1 11.3 ± 0.9 0.3 ± 0.1

5.4.3. Recycled powders Vs side surface

The side surfaces of the SLM fabricated components had more particles

sintered to them. Hence a comparison of the surface chemistry of the particles

used for fabrication and the surface chemistry of the side surface was

examined to study the difference in their surface chemistry. Figure 5-24 shows

the surface morphologies of the powders used to fabricate the part and the side

surface of the part fabricated by SLM. From the figure it can be witnessed that

the feed particles are partially melted and bound to the laser melted zones.

Thus, the side surface of the SLM sample is contained with the feed particles

and the laser melted zones.


Figure 5-24 Surface morphology of recycled powders used to fabricate the part (a) and the

side surface with partially sintered particles (b).

The relative atomic percentage of the elements detected during XPS

measurement of the feed powder and the side surface is given in Table 5-5.

Compared to the powders, the side surface showed a relatively high degree of

carbon and nitrogen contamination due to their high surface area. However,

another possible reason for the increased carbon and nitrogen levels might be

due to the formation of more metal carbide and nitride on the SLM surface. A

small contribution to the carbon may also be possible from the carbon tape

used to stick the Ti6Al4V powder particles for characterisation. Since the

surface and the powders had varying levels of contamination, the ratio of metal

to oxygen was used to interpret their surface chemistry.

Table 5-5 Relative atomic percentage of elements detected during XPS characterisation of

Ti6Al4V powders and side surface. The calibration error during peak integration is ± 10%.

Sample

Type



Recycled

Powder 9.2 ± 0.6 23.8 ± 1 2.1 ± 0.2 50.5 ± 0.6 14.2 ± 1 0.3 ± 0.1

Side

Surface 5.4 ± 0.5 25.8 ± 2.1 3.2 ± 0.8 49.8 ± 1.6 15.8 ± 0.9 0

On comparing the metal to oxygen ratio of the major alloying elements Ti and

Al, it was observed that the Ti:O and Al:O ratios of the powders and the side

surface were significantly different with varying levels of surface oxide

compositions (Figure 5-25). The ratio of Ti:O for the Ti6Al4V powder was

lower than the ratio of the side surface. This implies that on laser processing of

the Ti6Al4V powders, the ratio of Ti:O increases. In contrast, the Al:O ratio


observed for the powders were higher than the ratio witnessed for the side

surface.

Figure 5-25 Elemental ratio of titanium and aluminium to oxygen in the recycled powder and

the side surface.

These clearly show that although a significant amount of particles are partially

melted and bound to the side surface, the surfaces do not exhibit the surface

chemistry of the feed powders. Since the presence of both SLM melted zones

and partially melted powders are observed on the surface, its surface chemistry

is most possibly contributed to by both of these factors. Also, controlling the

surface chemistry of a SLM surface with both partially sintered particles and

laser melted zones can be difficult.

5.5. Summary

SLM fabricated Ti6Al4V samples were observed to be rough and porous due

to partially melted particles on the surface. Disappearance of carbon on etching

the surface by approximately 4 nm revealed most of the carbon observed on all

three surfaces was likely due to contamination. As revealed by the depth

profiling, the present study showed differences in the surface chemistry of

SLM-AF, SLM-MP and FGD-MP with respect to the distribution of elements

and its composition on the surface. Vanadium was not observed on the SLM-

AF surface and hence the surface oxide film of SLM-AF was made of oxides

of titanium and aluminium. Due to the existence of vanadium, the surface

0.28

0.18

0.32

0.11

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

Ti:O Al:O

Me

tal to

th

e o

xyg

en

ra

tio

Recycled Powder

Non-laser Scanned Area


oxide film of SLM-MP and FGD-MP was made of all three alloying elements

(Ti, Al and V). Although SLM-AF showed a high oxide film thickness, this

could have been due to the effect of partially melted particles on the surface. A

SLM fabricated part with a smooth surface will be ideal to determine its oxide

layer thickness accurately using XPS. On comparing the impact of laser power

on the surface chemistry, when skin scanning was performed, a possible

increase in the surface oxide layer and segregation of aluminium on the surface

was observed. The study on the top and side surfaces showed differences in

their surface chemistries due to the varied elemental distributions. Thus, the

SLM fabricated parts showcased a non-homogeneous surface chemistry on its

surfaces and the laser power was observed to influence the surface chemistry

of the SLM fabricated parts. Due to this non-homogeneous surface chemistry,

surface modification is required to prevent the release of metal ions by which

corrosion resistance and cytocompatibility of the SLM fabricated Ti6Al4V

components can be enhanced for biomedical applications.

Chapter 6: Surface Modification Using SAMs 112

6. SURFACE MODIFICATION USING SAMs

6.1. Introduction

In chapter 5, the SLM fabricated surfaces were shown to have a non-

homogenous surface chemistry and require surface modification. In this

chapter, surface modification by the adsorption of SAMs and in vitro stability

of the attached phosphonic acid monolayers formed on SLM fabricated

surfaces are described. The impact of surface roughness on the stability of the

chemisorbed monolayers is also presented.

6.2. SAM Attachment

Previous studies on the attachment of phosphonic acid monolayers have been

performed mostly on polished surfaces [53–56,58,137,144,189] with only a

very few having been performed on rough surfaces [149]. For biomedical use,

some applications require a smooth surface (such as stents) and some require a

porous surface to promote cell adhesion and tissue integration/regeneration[4].

Prior examples of the attachment of monolayers to an SLM fabricated surface

have not been reported. Hence, the attachment of 16-PhDA phosphonic acid

SAMs (with COOH termination/functionality) to a rough SLM as-fabricated

Ti6Al4V surface (SLM-AF) and a smooth mechanically polished (SLM-MP)

surface was performed. The surface roughness (Ra) of the SLM-MP after

polishing was measured to be 590 ± 13.5 nm. In vitro stability of phosphonic

acid monolayers on these rough and smooth surfaces and the impact of surface

roughness on SAM’s stability were investigated.

Since one batch of the samples was rough (SLM-AF), the commonly used

techniques to characterise SAMs including atomic force microscopy (AFM)

and ellipsometry could not be employed in this study. Also the presence of

irregular pores on the mechanically polished surfaces affected AFM

measurements. Hence, XPS was used as the main characterisation technique

along with contact angle goniometry to confirm the attachment and stability of

the monolayers.


6.2.1. Surface chemistry

Figure 6-1shows the XPS spectra of the SLM-AF and SLM-MP Ti6Al4V

surfaces before and after 16-PhDA SAM coating. It can be observed that there

is a significant change in the carbon 1s peak at 285 eV after surface

modification using SAMs for both SLM-AF and SLM-MP. This increase in

carbon is consistent with the presence of the 16 carbon atoms of 16-PhDA.

Also after surface modification, a significant amount of phosphorous was

observed on the SAM-modified surface for both SLM-AF and SLM-MP.

Although a ratio of 16:1 would be expected, the observed C:P atomic ratio was

approximately 13:1 for SLM-AF and 12.4:1 for SLM-MP. This may be due to

peak fitting uncertainties (calculated error was ± 10%). Also, since XPS

measurement is sensitive to surface contaminants, it is believed that

contamination of the sample might have affected the ratio. For example,

contamination of the surface due to oxygen and nitrogen based contaminants

can alter the atomic percentage of elements. A small difference in the C:P ratio

due to surface contaminants has been reported in literature [190]. However, it

should be noted that the atomic percentage of phosphorous is much less than

that of carbon, and is consistent with the presence of only one phosphorous

atom in 16-PhDA.

The observed low concentrations of aluminium (SLM-MP) and nitrogen (for

SLM-AF) on the Ti6Al4V surfaces before SAM attachment further reduced

after SAM formation. The observed nitrogen might be due to contamination of

the surface due to atmospheric contaminants such as oils. Also the

concentration of titanium and oxygen observed before SAM attachment

decreased after surface modification, since the limited penetration depth of

XPS is now sampling less of the underlying substrate. The detection of

titanium and oxygen through the 16-PhDA layer confirms that this layer is thin

(i.e. less than the typical XPS sampling depth of 5-10 nm). A similar decrease

in the bulk material’s composition was reported in previous literature after the

assembly of SAMs [144]. All observed changes in the surface chemistry of the

SLM-AF and SLM-MP surfaces are consistent with previously reported

literature for the formation of SAMs on the SLM fabricated Ti6Al4V surfaces


[37]. Thus the XPS data is consistent with the coating of 16-PhDA SAMs on

both the SLM-AF and SLM-MP surfaces.

Figure 6-1 XPS spectra before and after surface modification of a) SLM-AF and b) SLM-MP

by adsorption of a 16-PhDA monolayer.

6.2.2. Surface wettability

The contact angle formed by a liquid at an interface can provide a measure of

surface wettability, SAM order and uniformity [55]. Figure 6-2 shows the

surface wettability of SLM-AF and SLM-MP surfaces after different surface

treatments. It can be observed that the contact angle reduced after surface

cleaning. The high contact angle value before surface cleaning is likely due to

the presence of contaminants on the surface. The important and problematic

contaminants are hydrocarbons (oils from pumps and skin) and impurities on


the material surface. These contaminants are mainly due to manual handling

and exposure to the atmosphere [1,132].

On cleaning these contaminants were effectively removed from the surface

thus yielding lower contact angles (due to high surface energy) for both SLM-

AF and SLM-MP surfaces. After surface modification using SAMs, the contact

angle reduced further. This was because the employed SAMs were

hydrophilic, possessing a carboxylic acid (COOH) group. This change in the

contact angle further supports the formation of a monolayer on both the SLM-

AF and SLM-MP surfaces. Differences between the measured contact angles

for the two surfaces were most likely due to the effect of the different surface

roughnesses of SLM-AF (17.6 ± 1.66 µm) and SLM-MP (0.59 ± 0.01 µm).

Due to the presence of voids in certain areas of the SLM-AF surface,

measurement of contact angles on those areas were impossible. Hence, an

alternate area (without voids) was selected on the sample for wettability

measurement.

Figure 6-2 Static water contact angle measurements on SLM-AF and SLM-MP surfaces, after

cleaning and SAM attachment.

Although both substrates were modified with SAMs on the same time, in

addition to their roughness profile, their varying levels of surface

contamination could affect their surface wettability. Previous literature

reported contact angle values ranging from 45° to 62° for the assembly of

69

43

30

78

45

21

0

10

20

30

40

50

60

70

80

90

Before Cleaning After Cleaning After SAM attachment

Co

nta

ct

An

gle

(°)

SLM-AF

SLM-MP


hydrophilic phosphonic acid SAMs [53,56,153,191]. However, the contact

angle observed in this study was lower than the values reported in literature.

This difference in the contact angle can be attributed to the monolayer

arrangement and the cleanliness of the substrate [146]. Also, surface roughness

pattern can affect the monolayer arrangement and the surface wettability.

Tosatti et al. [149] demonstrated the surface wettability of SAM modified

titanium surface to vary with their varied roughness patterns. In their study, the

rough surface (Ra = 2.5 ± 0.2 µm) showed a advancing contact angle of 150°

for methyl terminated SAM; whereas, the smooth surface (Ra = 3 ± 0.3 nm)

depicted 110°. Thus surface roughness patter can affect the contact angle

measurements. However, since the phosphonic SAMs used in this study bears

–COOH functionality, a hydrophilic group, the surface remained hydrophilic.

This confirms the surface modification of SLM-AF and SLM-MP Ti6Al4V

surfaces using 16-PhDA SAMs.

6.3. Stability Studies


Figure 6-3 shows the high resolution XPS spectra for the phosphorous 2p

region of the samples soaked in the buffer solution for different time intervals.

It can be witnessed from the spectra that the metal phosphonate peak at 133.3 ±

0.2 eV retained its position at the same binding energy for the whole duration

of immersion in the buffer solution. Bhure et al. [55] reported the phosphorous

2p peak to be centred at 133.2 eV for 14 days during their stability studies

(when exposed to air). Similarly, in this study, the P 2p peak was observed at

133.3 ± 0.2 eV throughout the study period. However, the intensity and

relative atomic composition of P 2p peak decreased over the course of the

experiment especially after four weeks, showing desorption of monolayers

from the Ti6Al4V surface. Although the desorption of monolayers was noted

to occur after four weeks, a small amount of monolayer was observed at the

end of six weeks. On comparing the XPS P 2p spectra of SLM-AF and SLM-

MP, there was no significant difference in the stability of 16-PhDA SAMs on

these surfaces. This shows that the roughness pattern of the SLM-AF did not

affect the stability. However, it should be noted that the peeling of partially


melted particles after SAM attachment can affect the monolayer assembly and

may lead to the formation of islands without monolayers.

Figure 6-3 XPS spectra for the in vitro stability of SAMs on SLM-AF (a) and SLM-MP (b)

surfaces after immersing in Tris-HCl buffer solution for various time intervals.


A gradual increase in the contact angle was noted after immersing the samples

in Tris-HCl buffer solution (although this change has low significance

compared to the standard deviations of individual measurements). This may be

attributed to a small amount of desorption (Figure 6-4) leading to the change in

the assembly pattern of the monolayers. It can be noted from the figure that the

contact angles measured for the SLM-AF samples had a high standard

deviation when compared to the SLM-MP samples. This may be due to the

surface roughness of the SLM-AF samples. The slight increase with time for

the SLM-MP sample was observed, however this was not significant. The

effect of roughness also makes these results difficult to interpret; however, the

surfaces of both SLM-AF and SLM-MP remained highly wettable. Bhure et al.

[55] studied the stability of phosphonic acid SAMs on cobalt-chromium

surfaces by exposing it to the atmosphere and reported that the SAMs were

stable for the duration of their study i.e. 14 days. Kanta et al. [192] studied the

stability of phosphonic acid SAMs by exposing the SAM coated titanium

surfaces to various conditions including the exposure to UV light (energy of

10.4 J/cm2 at sample surface) for 180 minutes and heat treatment (100 °C) for

3 hours and plasma treatment with a power of 100W and frequency of 13.56

MHz. Their studies concluded that UV, heat and plasma treatment decomposed


the alkyl chain leading to the degradation of SAMs. However, there were no

previous studies reporting the long-term stability (6 weeks) of the 16-PhDA

monolayers assembled on to SLM fabricated Ti6Al4V surfaces (rough and

smooth) under in vitro condition. Hence, it is believed that this is possibly the

first study showing their stability under in vitro conditions.

Figure 6-4 Static water contact angle measurements on SLM as-fabricated (SLM-AF) and

SLM-MP surfaces for different immersion time intervals in Tris-HCl buffer solution. “Before

SM” refers to samples before surface modification using SAMs and “Week 0” measurements

made after surface modification.

6.4. Summary

The as-fabricated SLM surface (SLM-AF) and mechanically polished SLM

surface (SLM-MP) were modified using 16-PhDA monolayers. XPS and

contact angle measurements were consistent with the formation of monolayers.

The surface roughness of the SLM-AF samples did not affect the monolayer

formation significantly. 16-PhDA monolayers were found to be stable for over

28 days on both SLM-AF and SLM-MP surface before its desorption.

However, a small amount of phosphorous was still observed on both SLM-AF

and SLM-MP surface after 42 day time interval. The rough nature of the SLM-

AF surface did not have a significant effect on the stability of the attached

monolayers. It should be noted that XPS produced comparatively lower signals

for as-fabricated SLM-AF surface compared to the mechanically polished

SLM-MP surface. The rough nature of the SLM-AF surface was also observed

0

10

20

30

40

50

Before SM Week 0 Week 1 Week 2 Week 4 Week 6

Co

nta

ct

An

gle

(d

eg

ree)

SLM-AF

SLM-MP


to affect the wettability measurements leading to scattering of data. Hence the

use of a mechanically polished surface would be beneficial for improved XPS

intensity and better surface wettability measurements.

Chapter 7: Functionalisation of SAMs 120

7. FUNCTIONALISATION OF SAMs

7.1. Introduction

On studying the in vitro stability of phosphonic acid monolayers,

functionalisation of these monolayers with Paracetamol as a model drug was

performed. In this chapter, the results obtained using XPS and surface

wettability measurements for the adsorption and functionalisation of SAMs on

SLM fabricated surface are discussed.

7.2. SAM attachment

As discussed in previous chapters, the SLM-produced surfaces are porous due

to partially melted particles. These particles could detach from the surface and

can cause acute and chronic effects. If these particles detach after monolayer

assembly, those areas will be left without drug attachment since there will be

no underlying monolayer to functionalise. XPS was also observed to be

sensitive to rough surfaces and thus reducing the signal. Hence, the SLM

fabricated surfaces were mechanically polished on one of the surfaces to

remove the partially-sintered particles and to attain a smooth surface before

surface modification with SAMs and functionalisation with a drug. The

average surface roughness (Ra) of the mechanically polished surfaces obtained

using a surface profilometer was 590 ± 13.5 nm.


Figure 7-1 shows the survey spectra obtained from the mechanically polished

Ti6Al4V surface before and after surface modification using 16-PhDA SAMs.

The spectra clearly indicate the introduction of a metal phosphonate peak at

133.3 eV showing the surface has been modified with 16-PhDA monolayers.

This metal phosphonate peak is due to the formation of Ti-O-P bonds after

SAM attachment [56,148]. This has further been justified by an increase in the

intensity of carbon and a reduction in the intensity of oxygen, titanium and

aluminium after surface modification (due to the limited penetration depth of

XPS measurements, less of the underlying metal and oxide is detected).


Figure 7-1 Survey spectra showing the change in surface chemistry of a Ti6Al4V surface after

SAM attachment. (a) Control surface without SAM coating and (b) After 16-PhDA SAM

coating.

Table 7-1 shows the relative atomic composition obtained for the Ti6Al4V

surface before and after surface modification. For a surface modified using 16-

PhDA, the expected ratio of C:P is 16:1 and in the current study, it is observed

to be nearly the same (i.e. 15.6:1). The small variation between the actual and

the expected ratio is within the instrument/calibration error (± 10%) limit. High

resolution spectra obtained for the phosphorous 2p region showed the

formation of a metal-phosphonate bond at 133.3 eV confirming the

modification of the surface with monolayers [56].

Table 7-1 Relative atomic percentages of the elements detected in XPS. The calibration error

during peak integration in XPS is ± 10%.

Sample Relative Atomic Percentage

C O N Ti Al V P Cl S

Control 21.8 ±

1 54.5 ±

1.2 0.2 ±

0.5 18.2 ±

0.5 4.7 ±

0.4 0.6 ±

0.1 0 0 0 16-PhDA

SAM 51.1 ±

1.3 32.9 ±

0.8 2.0 ±

0.3 8.7 ±

0.6 1.8 ±

0.4 0.3 ±

0.1 3.3 ±

0.3 0 0

The C:P ratio obtained in this chapter for 16-PhDA SAMs was closer to the

actual than the C:P ratio of SLM-AF (13:1) and SLM-MP (12:1) obtained in

the previous chapter. One of the possible reasons for this difference is the

concentration of carbon. An increase or decrease in the carbon concentration

0 100 200 300 400 500 600

Rela

tiv

e I

nte

nsit

y (

Arb

itra

ry U

nit

s)

Binding Energy (eV)

P 2P, P

C 1S

O 1S

Ti 2p

Ti 3s

Al 2P b

a

N 1S


can affect the C:P ratio [190]. Carbon is present in the atmosphere as

contaminant. Similarly, contamination is also possible due to oxygen and

nitrogen, leading to variations in the C:P ratio. Hoque et al. [190] also

observed a difference in C:P ratio due to surface contamination.


The average static contact angles obtained from the surface before and after

surface modification with SAMs were 24° ± 1.3° and 35 ± 3.1° respectively.

The contact angles can be observed to increase after SAM attachment showing

a change in the surface wettability. A similar trend of a change in contact angle

after surface modification using SAMs was previously reported [37]. This

change is attributed to confirm the surface modification, since surface

wettability is highly specific to the surface chemistry. However, these

wettability values are different to the values reported for ‘before surface

modification’ and ‘after SAM attachment’ in chapter 6. As mentioned in

chapter 6, a small disorder in the assembly of monolayers can offer varied

surface wettability/contact angles. Since surface profiling of the SAM attached

Ti6Al4V surfaces using an AFM was not performed in this study (due to the

rough nature of the sample), it is difficult to qualify the monolayer

order/arrangement pattern.

Mahapatro et al. [58] reported a significant change in contact angle before

cleaning (94.3°) and after cleaning (< 3°) of 316L SS surface (Ra = 0.2 ± 0.1

µm). Also, their varied surface chemical treatment provided them with varied

contact angles. On treating bare metal surface with 70% ethanol, acetone and

40% nitric acid separately, their contact angle was 64°. However, this reduced

to < 3° on plasma treatment of the chemically treated surface. This shows that

varied chemical treatment will provide varied contact angles. Faucheux et al.

[48] reported contact angles between 48° and 62° for COOH SAMs for silica

and glass substrates (Ra 0.3 nm); whereas, the contact angle observed in the

current work was low. This can be due to various factors including surface

roughness and monolayer arrangement pattern, varying levels of surface

contamination, differences in the purity and pH of water used to measure the

contact angle. However, since 16-PhDA SAMs are hydrophilic, possessing a


carboxylic acid (COOH) terminal group, the surface remained highly wettable

after surface modification.

7.3. SAMs Functionalisation with Paracetamol


After studying the stability of 16-PhDA monolayers on the SLM fabricated

Ti6Al4V surface, a sample drug (Paracetamol) was used to functionalise the

SAMs. Paracetamol was only used as a model drug to test the reaction

mechanism. As described in the methodology, the carboxylic acid group at the

terminal end of 16-PhDA SAMs was reacted with thionyl chloride (SOCl2) to

form an intermediate acid chloride compound with –COCl as the terminal

group. Other than SOCl2 to form an acid chloride, phosphorous trichloride

(PCl3) or phosphorous pentachloride (PCl5) can also be used. After forming the

acyl chloride, this group was then allowed to react with the hydroxyl (–OH)

group of Paracetamol to bind the drug covalently to the monolayers. The

whole reaction step was performed in an inert atmosphere to prevent the SOCl2

reacting with moisture in the atmosphere to form HCl that could hydrolyse the

–COCl to inactive –COOH. It should be noted that no base was used during

this reaction.

Scheme 7-2 shows the reaction schemes for the immobilisation of Paracetamol

to SLM fabricated Ti6Al4V surface. Figure 7-2 shows the survey spectra for

the Ti6Al4V sample before and after drug attachment. The corresponding

atomic percentages are given in Table 7-2. It can be clearly noted that there is a

significant change in the surface chemistry of the SAM-coated Ti6Al4V

surface after the immobilisation of Paracetamol. The ratio of C:P was 15.6 for

SAM coated surface and after the attachment of Paracetamol the ratio of C:P

was 41. Also, there is an increase in the atomic percentage of nitrogen. These

suggest the addition of Paracetamol to the SAM coated surface. This addition

of material is likely to be Paracetamol.


Scheme 7-1 Scheme of reactions to coat 16-PhDA monolayers to a Ti6Al4V surface and to

functionalise the monolayers with Paracetamol. Label description: hydroxylated Ti6Al4V

surface (1); 16-PhDA SAMs (2); 16-PhDA SAMs chemisorbed to Ti6Al4V surface (3);

reaction intermediate (4); Paracetamol immobilised to SAMs (5).

Figure 7-2 Survey spectra obtained using XPS for the SAM coated and Paracetamol coated

surfaces. The changes in the intensity of the detected elements before and after drug coating

indicate a change in the surface chemistry.


Table 7-2 Relative atomic percentages of elements detected by the XPS in SAM coated and

Paracetamol coated Ti6Al4V surfaces. The calibration error during peak integration in XPS is

± 10%.

Sample


C O N Ti Al V P Cl S

16-PhDA

SAM

51.1 ±

1.3

32.9 ±

0.8

2.0 ±

0.3

8.7 ±

0.6

1.8 ±

0.4

0.3 ±

0.1

3.3 ±

0.3 0 0

SAM +

Paracetamol 48.1 ± 1

35.8 ±

0.9

4.5 ±

0.4

7.8 ±

0.7

1.1 ±

0.3 0

1.2 ±

0.2

0.8 ±

0.1

0.9 ±

0.1

Due to the presence of nitrogen in Paracetamol, its concentration was noted to

increase; however, this is not a large increase since the drug molecule contains

only one nitrogen atom. The concentration of phosphorous (from 16-PhDA)

and the underlying metals (titanium, aluminium and vanadium) were observed

to decrease and this further confirms the functionalisation of SAMs with

Paracetamol. A very small contribution from chlorine and Sulfur was observed

after the attachment of Paracetamol and this is likely to be added from the

SOCl2 used in the intermediate step.

Carbon contamination is unavoidable as it is present as a contaminant in the

atmosphere and in the solvent THF. However, a small amount of nitrogen was

also observed on the Ti6Al4V sample as a contaminant in both the control and

SAM coated surfaces. The presence of these elements on the metal oxide

surface as contaminants has been reported in literature [45,56]. Hence the ratio

of carbon to the underlying metal composition of the surface has been used to

determine the adsorption of SAMs and Paracetamol to the metal surface.

Figure 7-3 shows the ratio of carbon and oxygen to the underlying metals

(titanium + aluminium and vanadium) and the ratio of carbon to oxygen for all

three samples (control, SAM attached and Paracetamol coated). It can be

observed that the carbon and oxygen concentration increased after SAM

attachment and functionalisation showing the attachment of the Paracetamol to

the tail group of 16-PhDA monolayers. A similar reduction in the base metal

composition after surface modification was reported previously [45].


Figure 7-3 Ratio of carbon and oxygen to its underlying metals and C/O ratio for control,

SAM coated and Paracetamol coated surfaces.

A comparison of the C 1s spectra obtained for the control, 16-PhDA SAM

coated and SAMs functionalised with Paracetamol is shown in Figure 7-4.

High-resolution spectrum of C 1s for the control sample was deconvoluted into

two components. Peaks formed at 285.3 eV and 286.2 eV were assigned to C-

C and C-O of hydrocarbon contaminants. Adsorption of hydrocarbon as a

contaminant to metal oxide surfaces have been discussed in previous literature

and the major source for this contaminant is adventitious carbon in air,

solvents (used for cleaning) and from hydrocarbons (containing proteins and

oils) due to manual handling [181,184,193,194].

The C 1s spectrum for the SAM coated Ti6Al4V surface was deconvoluted

into three components (Figure 7-4b). The peaks observed at 284.9 eV, 286.5

eV and 289.3 eV were assigned to C-C, C-O and C=O [45]. A significant

increase in the relative C-C peak intensity when compared to the control

surface was due to the 16 membered carbon chain [HOOC(CH2)15PO(OH)2] in

the 16-PhDA molecule. The C-O and C=O are also from the 16-PhDA

molecule, present in the head and tail group of the chain. The significant

change in the intensity of C-C, C-O and the introduction of C=O peak confirms

the modification of SLM fabricated Ti6Al4V surface with 16-PhDA

monolayers.

0

1

2

3

4

5

6

C/Metals (Ti+Al+V) O/Metals (Ti+Al+V) C/O

XP

S R

ati

os

Control

SAM only

SAM+Paracetamol


On characterising the Ti6Al4V surface functionalised with Paracetamol

(Figure 7-4c), the deconvoluted C 1s spectrum rendered four components at

284.7 eV, 285.8 eV, 286.9 eV and 289.1 eV. These peaks can be assigned to

C-C, C-O, C-N and C=O, respectively [45,144]. Comparing this to the SAM

coated surface, the peak intensity of C-C and C=O has changed. Also the

introduction of C-N peak at 286.9 eV was observed. This C-N peak is from the

amide group in Paracetamol.

Figure 7-4 High resolution spectra of carbon 1s region obtained for (a) control, (b) SAM

attached, (c) Paracetamol coated samples, (d) shows the C 1s region for Paracetamol powder

characterised using XPS.

Paracetamol powder in its pure from (as purchased) was characterised using

XPS to obtain the spectrum for C 1s region (Figure 7-4d). The deconvoluted

spectrum showed the existence of four components C-C (284.7 eV), C-O (286

eV), C-N (288.2 eV) and C=O (291.2 eV) [45,160]. These peaks can be

assigned to carbon atoms in hydrocarbon, hydroxyl, amide and ester groups of

Paracetamol. The XPS results obtained for Paracetamol coated Ti6Al4V

surface were in good agreement with the results obtained for the C 1s region of

Paracetamol powder. However, a small change in the peak binding energies

was noted and this could be due to errors in peak integration during

deconvolution (± 0.1 %) and might also be due to sample charging (± 0.3 eV)

of the Paracetamol powder.


The O 1s region spectrum obtained using XPS for the control sample (Figure

7-5a) was deconvoluted into two components, metal oxide at 530.5 eV and

oxygen atoms in O-C at 532.1 eV [156,183,195]. The metal oxide peak

observed at 530.5 eV is likely to be mostly due to titanium oxide; however, a

small proportion of aluminium and vanadium oxides is also possible due to the

presence of these elements in Ti6Al4V alloy. In agreement with the C 1s

spectrum, the O 1s spectrum also showed the existence of O-C species on the

control sample as contaminants.

On deconvoluting the O 1s spectrum for the 16PhDA SAM coated Ti6Al4V

surface, three components were obtained (Figure 7-5b). The peaks formed at

530.4 eV, 531.9 eV and 533.2 eV were assigned to metal oxides, O-C and

O=C, respectively [156,183,196]; however, there is also the possibility for a

small contribution from aluminium oxide for the peak observed at 531 eV. The

relative intensity of metal oxide peak at 530.4 eV for the SAM coated sample

can be observed to decrease and the O=C peak at 533.2 eV to increase when

compared to the control sample. The introduction of O=C at 533.2 eV should

mostly derive from the carboxylic group at the terminal end of the 16-PhDA

molecule. These observed changes compared with the control sample can be

attributed to the adsorption of 16-PhDA monolayers on to the SLM fabricated

Ti6Al4V surfaces.

After functionalising SAMs with Paracetamol, the deconvoluted O 1s XPS

spectrum rendered three components metal oxide (530.8 eV), O-C (532.1 eV)

and O=C (533.4 eV) and are shown in Figure 7-5c. Although the O-C and O=C

are expected only from Paracetamol, a small contribution from the underlying

monolayers is possible (see 7-5b). The XPS characterisation of Paracetamol

powder for the O 1s region rendered two components including O-C (531.7

eV) and O=C (533.1 eV) on deconvolution [183] as shown in Figure 7-5d. On

comparing the deconvoluted O 1s peaks obtained for the Paracetamol coated

surface with the O 1s peaks obtained for the Paracetamol powder, both the

peaks were in good agreement. However, there was a slight variation in the

binding energies. This further confirms the attachment of Paracetamol to the

16-PhDA adsorbed Ti6Al4V surface.


Figure 7-5 XPS spectra obtained for the oxygen 1s region for (a) control, (b) SAM attached,

(c) Paracetamol coated samples, (d) shows the O 1s region for Paracetamol powder

characterised using XPS.

Deconvoluted N 1s spectrum showed the presence of N-C and N-O at 400.3

eV and 402.6 eV respectively on the control surface (Figure 7-6a). Although

nitrogen is not expected on the Ti6Al4V surface (as it is not in the

composition), a small amount of nitrogen was observed. Similar to carbon,

nitrogen also has strong affinity towards metal oxides and their presence on

metal oxide surfaces could be from nitrogen containing carbon contaminants

[45]. Similar to the control sample, the SAM coated surface also showed the

existence of nitrogen at 400.1 eV (N-C) and 402.3 eV (N-O) after

deconvolution (Figure 7-6b) [181]. This could also be from contamination of

the surface as the peaks were observed similar to the control samples.

Furthermore, the 16-PhDA SAMs do not have nitrogen in their structure.

The deconvoluted N 1s spectrum for the Paracetamol coated Ti6Al4V surface

also showed two components, N-C at 400.1 eV and Ti-Ox-Ny at 401.9 eV

(Figure 7-6c). Although a small contribution to these peak from contaminants

is possible since the other samples (control and SAM coated) were shown to

have contaminants, the N-C peak is distinct and increased in relative intensity.


XPS characterisation of Paracetamol powder for the N 1s region showed a

single N-C peak at 400.3 eV (Figure 7-6d).

Figure 7-6 N 1s region for (a) control, (b) SAM attached, (c) Paracetamol coated samples, (d)

Paracetamol powder, characterised using XPS.


Static contact angles obtained for the control, 16-PhDA SAM attached and

Paracetamol coated surfaces are shown in Figure 7-7. Figure 7-8 shows the

contact angle values obtained graphically. It can be observed that the control

and 16-PhDA SAM deposited surfaces were highly wettable whereas the

Paracetamol coated surface was more hydrophobic. The wettability of the

control surface was due to the presence of metal oxides which impart a high

surface energy. Wettability of the 16-PhDA SAM coated surfaces was due to

the presence of carboxylic acid in their terminal group. A similar increase in

contact angle after SAM modification was reported in literature [54,58]. The

difference in the contact angle is due to the change in surface chemistry.

Control Ti6Al4V surface will present hydroxylated surface oxide layer, and

COOH-SAM modified Ti6Al4V surface will also present hydroxylated

surface. However, it is not always possible to get a very well packed SAMs.

Even if there is a small change in the ordering, this will affect the contact

angle. Hence this might be the reason for the difference in contact angle before

and after SAM attachment. Since the SAM coated surface was functionalised

with Paracetamol, it would be expected that a methyl group (-CH3) is exposed

as the terminal group leading to a lower energy surface. In agreement with


this, the measured contact angles showed the Paracetamol coated surface to be

more hydrophobic with a contact angle higher than 90°. Mahapatro et al. [197]

reported a change in the surface wettability from hydrophilic to hydrophobic

surface after the attachment of Ibuprofen to SAMs. Thus the change in surface

wettability is consistent with the functionalisation of Paracetamol to 16-PhDA

SAMs coated on the SLM fabricated Ti6Al4V surfaces.

Figure 7-7 Contact angles obtained for SLM fabricated Ti6Al4V surfaces after cleaning (a),

16-PhDA SAM attachment (b) and Paracetamol attachment (c).

Figure 7-8 Static water contact angle obtained for Control, SAM coated and Paracetamol

coated Ti6Al4V samples.

Although the changes in the surface chemistries of Ti6Al4V surfaces revealed

the functionalisation of Paracetamol to SAMs, the use of Paracetamol as a

model drug may not be advisable. This is mainly because of two reasons. The

first being, Paracetamol does not have a distinct element that could easily be

characterised. The elements constituting Paracetamol are carbon, nitrogen,

24

35

95

0

20

40

60

80

100

120

Control SAM Attached SAM+Paracetamol

Co

nta

ct

An

gle

(°)

Sample Type


oxygen and hydrogen. All of these elements are present in the atmosphere as

contaminants. Hence, the use of these elements to confirm the reaction may

lead to false results. The second reason is, Paracetamol is a phenolic

compound. On functionalising Paracetamol to COOH SAMs, a phenolic ester

is formed. In general phenolic esters will hydrolyse readily in the presence of a

hydroxyl group to form an alcohol and carboxylic acid [172]. Hence, moisture

in the atmosphere can hydrolyse the drug. Thus they have poor oxidative

stability and will also show a rapid release under physiological conditions.

Thus Paracetamol may not be the best choice of drug for these reasons.

Pyrene derivatives such as hydroxy-pyrenes, amino-pyrenes can be used to

study the functionalisation since they have a better hydrolytic stability than

phenolic esters. In addition to this, they are fluorescent compounds and hence

they could be easily characterised using a fluorescence microscopy. However,

these compounds mainly constitute carbon and nitrogen, which are present in

atmosphere. This could limit the use of characterisation equipment such as

XPS.

Fluoroquinolone compounds such as Ciprofloxacin®, Levofloxacin®,

Ofloxacin® may be a better choice than Paracetamol and Pyrene derivatives

since they have a distinct fluorine atom (not usually present in air as

contaminant) attached to the central benzene ring that could easily be

characterised in FTIR and XPS. Also, most of the fluoroquinolones exhibit

fluorescence property, which could then be used for characterisation in

fluorescence microscopy to determine the surface coverage of the compound

[198]. Since these compounds being antibiotic, simple biological tests can be

performed to confirm the drug attachment and its activity. Thus a careful

consideration is required when selecting a model drug.

7.4. Summary

Changes observed in the surface chemistry of the Ti6Al4V surface at each

stage through XPS characterisation and surface wettability measurements are

consistent with the functionalisation of monolayers with Paracetamol. This

chapter showed the successful immobilisation of Paracetamol to 16-PhDA

SAMs attached to a SLM fabricated Ti6al4V surface. However, nitrogen alone


cannot be used as a key element to confirm Paracetamol attachment, since it is

present on the surfaces as a contaminant and may affect the result. Also,

phenolic esters hydrolyse readily. Hence, the use of compounds with distinct

elements that is unlikely to be present in the atmosphere as contaminants

should be used to prevent contamination affecting the results of future studies.

Chapter 8: Immobilisation of Ciprofloxacin® 134

8. IMMOBILISATION OF CIPROFLOXACIN®

8.1. Introduction

In this chapter, immobilisation of an anti-bacterial drug, Ciprofloxacin®, to

SAMs using the previously stated procedure for Paracetamol is described.

Drug quantification studies, in vitro and oxidative stability of the drug, drug

release profile and antibacterial susceptibility of Ciprofloxacin® adsorbed to a

SLM fabricated Ti6Al4V surface are presented.

8.2. SAM attachment


XPS survey spectra for the control octadecylphosphonic acid (ODPA) SAM

coated and 16-phosphanohexadecanoic acid (16-PhDA) SAM coated Ti6Al4V

samples are shown in Figure 8-1. Peaks for the major alloying elements Ti and

Al are clearly visible from the spectra. Since the Ti6Al4V surface

predominantly contained oxides of Ti and Al, the amount of V observed on the

surface was considerably less (<0.3%). Hence the V peak is not clearly visible

in the spectra. The relative atomic percentages of the detected elements before

and after surface modification are shown in Table 8-1.

Figure 8-1 Survey spectra shows the change in the intensity of C 1s peak before and after

SAM attachment and the introduction of P 2p peak after SAM attachment thus confirming the

attachment of 16-PhDA SAMs and ODPA SAMs to the Ti6Al4V surface.


Table 8-1 Relative atomic percentage of the elements detected by XPS after different surface

treatments. The calibration error during peak integration in XPS is ± 10%.

On comparing the spectra obtained for 16-PhDA and ODPA SAM coated

samples with the control spectrum, the intensity of C 1s peak is significantly

higher for the SAM coated samples. Also the intensities of O 1s and Ti 2p

Detected

Elements

Sample Type

Control 16-PhDA

SAM

Coated

16-PhDA

SAM + Drug

Coated

ODPA

SAM

Coated

ODPA

SAM +

Drug

Coated

Al 2p

Mean 4.19 2.39 1.24 2.44 2.26

S.E ± 0.4 ± 0.2 ± 0.1 ± 0.1 ± 0.2

Ti 2p

Mean 17.72 7.57 4.68 5.7 4.55

S.E ± 0.6 ± 0.4 ± 0.3 ± 0.4 ± 0.2

V 2p

Mean 0.63 0 0 0 0

S.E ± 0.1 0 0 0 0

C 1s

Mean 22.61 53.68 60.57 54.18 56.87

S.E ± 1 ± 0.5 ± 0.7 ± 0.5 ± 0.7

O 1s

Mean 50.93 31.05 24.84 31.87 28.27

S.E ± 1 ± 0.4 ± 0.8 ± 0.5 ± 0.6

N 1s

Mean 1.19 0.82 4.1 1.4 2.38

S.E ± 0.1 ± 0.1 ± 0.5 ± 0.2 ± 0.1

F 1s

Mean 0 0 1.5 0 0.63

S.E 0 0 ± 0.1 0 ± 0.1

P 2p

Mean 0 3.51 2.29 3.02 2.38

S.E 0 ± 0.2 ± 0.1 ± 0.21 ± 0.2

Cl 2p

Mean 0 0 0.41 0 0.28

S.E 0 0 ± 0.1 0 ± 0.1

S 2p

Mean 0 0 0.37 0 1.35

S.E 0 0 ± 0.01 0 ± 0.1

Si 2p

Mean 2.73 1.98 0 1.39 1.03

S.E ± 0.8 ± 0.6 0 ± 0.6 ± 0.5


peaks were lower for the SAM coated samples compared to the control sample.

The presence of phosphorous from 16-PhDA and ODPA SAMs was confirmed

by the existence of P 2p (133.3 eV) and P 2s (193.1 eV) peaks on the SAM

coated surfaces but not on the control surface. The atomic percentage of

phosphorous was much lower than that of carbon, consistent with the presence

of only one phosphorous atom in 16-PhDA. The obtained C:P atomic ratio for

the Ti6Al4V surface coated with 16-PhDA SAMs was 15.3:1; however, a ratio

of 16:1 would be expected. This small difference in the ratio lies within the

error limit of ± 10% due to the integration the P 2p peak.

Similarly, for the ODPA SAM coated surface a ratio of 18:1 would be

expected and the obtained ratio was 17.9:1. These values are in good

agreement with the literature previously reported for the modification of metal

surfaces with 16-PhDA and ODPA SAMs [54,55,148]. Thus, the XPS results

confirmed the successful deposition of 16-PhDA and ODPA SAMs on the

SLM fabricated Ti6Al4V surface. Carbon contamination of the control surface

was witnessed by XPS and this could be from the adventitious carbon present

in the atmosphere [183]. A small amount of silicon was observed on the

Ti6Al4V surface at all stages. This may be attributed to the contamination of

the samples during polishing with silicon carbide grits or during further

processing or characterisation.


The contact angle formed at an interface is sensitive to the molecular structure

of the surface and renders a measure of hydrophobicity or hydrophilicity of the

surface. These can also be attributed to the SAM arrangement and uniformity

of the surface coverage [55]. Hence, static water contact angle measurements

were performed using a contact angle goniometer to determine the change in

surface chemistry before and after surface modification with SAMs. Figure 8-2

shows the static water contact angles measured for the control, 16-PhDA SAM

coated and ODPA SAM coated samples.

The contact angle value obtained for the control sample (after cleaning)

showed the hydrophilic nature of the Ti6Al4V surface. After surface


modification with SAMs, this was observed to change. The 16-PhDA SAM

coated surface exhibited a hydrophilic nature due to the presence of a

carboxylic acid (-COOH) group in its terminal end (figure 8-2). On the other

hand, the ODPA SAM coated surface rendered a hydrophobic surface due to

the presence of a methyl group (-CH3) in its terminal group. Previous literature

showed the contact angle for ODPA SAM coated surface between 90° – 120°

[54,55,190]. The obtained contact angles for the ODPA SAMs were in

agreement with the previously reported values.

On comparing the obtained contact angles with chapter 7, the contact angle for

the Ti6Al4V substrate after cleaning was observed to be the same; however,

the contact angle of 16-PhDA SAM coated surface was different. As discussed

in previous chapters, the deviation in contact angle can be due to various

factors including SAM arrangement, surface cleanliness, surface roughness

and more importantly, the surface contamination. In agreement with the XPS

data, the change in surface wettability also confirmed the adsorption of 16-

PhDA and ODPA SAMs to the Ti6Al4V surfaces.

Figure 8-2 Static water contact angle showing a change in the wettability of the surface after

coating the surfaces with monolayers.

24

41

100

0

20

40

60

80

100

120

Control 16-PhDA SAM coated ODPA SAM Coated

Co

nta

ct

An

gle

(°)

Sample Type


8.3. Functionalisation of SAMs with Ciprofloxacin®


The relative atomic percentages of elements detected before and after surface

modification with SAMs and Ciprofloxacin® have been shown in Table 8-1.

From the table, changes in the relative atom percentage of elements were

observed for both 16-PhDA and ODPA SAMs after functionalising with

Ciprofloxacin®. However, 16-PhDA SAMs functionalised with

Ciprofloxacin® showed a more significant change than the ODPA SAMs

functionalised with the drug. After functionalising 16-PhDA SAMs with

Ciprofloxacin, a C:F ratio of 33:1 would be expected and the obtained results

were approximately 40:1. This difference in the ratio is believed to be due the

± 10% error in the peak integration and surface contamination before and

during surface characterisation. Also, the carbon contamination of the drug

coated surface might also be a contributing factor.

Although a small increase in the relative atomic percentage of C, N and a

decrease in the relative atomic percentage of Al, Ti, O and P was observed for

ODPA SAMs functionalised with the drug, this is not considered to be

significant (Table 8-1). This is because the atomic percentage of fluorine

obtained for ODPA SAMs functionalised with the drug was less than half of

the relative atomic percentage of F for 16-PhDA SAMs functionalised with the

drug. A C:F ratio of 35:1 would be expected for ODPA SAMs functionalised

with Ciprofloxacin®; however, the obtained ratio was nearly 90:1. This is

possibly because the tail group of the ODPA SAMs were not functionalised

with the drug due to the unavailability of a reactive group (such as –OH, -

COOH and -NH-) in ODPA SAM’s tail end. This shows that the small change

in the relative intensities and relative atomic percentage for the ODPA SAMs

after treating with Ciprofloxacin® may be due to the physisorbed drug

molecule on the surface.

Figure 8-3 shows the XPS spectra of the ODPA SAM coated Ti6Al4V before

and after functionalisation with Ciprofloxacin®. Whilst functionalisation of

methyl (-CH3) terminated SAMs with Ciprofloxacin® are not possible, a small


change in the surface chemistry of the ODPA SAM coated surface after

functionalisation can be observed from the spectra. A change in the intensity of

C, Ti, N and Al was observed along with the addition of fluorine at 688.2 eV.

These changes can be due to the physisorption of Ciprofloxacin® to the ODPA

SAM coated surface. There is also the possibility for the drug to displace some

ODPA and bond directly to the metal surface or perhaps fill the gaps in the

monolayer by binding to the surface. Similar to the functionalisation of 16-

PhDA SAMs with Ciprofloxacin®, the ODPA SAMs functionalised with the

drug also showed the physisorption of chlorine and Sulfur to its surface from

the intermediate step. The other potential binding site for Ciprofloxacin® is

through the amine group [171]. However, as previously discussed in the

methodology, since no base was used in this reaction, the reaction is believed

to occur through the carboxylic group to form acid anhydride rather than an

amide.

As the functional group of both Ciprofloxacin® and the 16-PhDA SAMs is a

carboxylic group (-COOH), they will not react with each other under normal

condition. Hence in order to provide favourable reactive conditions, the 16-

PhDA SAM coated surfaces were treated with thionyl chloride (SOCl2) in an

inert atmosphere to form an acid chloride. Now, the reactive –COOH group in

Ciprofloxacin® will readily react with the –OCl group of SAMs to immobilise

Ciprofloxacin® to SAMs (depending on the reaction conditions the amine

group in the Ciprofloxacin® molecule may also react with the SAMs).

However this is not possible with the ODPA SAMs since they do not have a

reactive terminal group.

Figure 8-4 shows the XPS survey spectra for 16-PhDA SAMs coated Ti6Al4V

surface functionalised with Ciprofloxacin®. The spectra clearly indicate a

change in the surface chemistry after functionalising the 16-PhDA SAM

coated surface. The intensity of carbon at 285 eV was observed to increase

after the drug attachment compared to the 16-PhDA SAM coated surface. This

increase is due to the presence of carbon in Ciprofloxacin®. The intensity of

nitrogen at 400.1 eV was also observed to increase due to the presence of N


(one secondary amine and 2 tertiary amines) in the chemical structure of

Ciprofloxacin®.

Figure 8-3 XPS survey spectra before (bottom) and after (top) functionalisation of a ODPA

SAM coated surface.

After functionalising the 16-PhDA SAMs with the drug, an addition of fluorine

(at 687.6 eV) to the SAM coated surface was observed. This fluorine is from

fluorocarbon present in the drug molecule. A small amount of silicon observed

on the SAM coated surface was noted to further decrease after the drug

attachment. Further to these changes, the decrease in the intensity of Al 2p, Ti

2p, O 1s and P 2p after drug attachment also confirmed the successful

functionalisation of 16-PhDA SAMs with Ciprofloxacin®. A very small

amount of sulfur in its oxidised form at 168 eV and chlorine at 193.9 eV was

observed. These sulfur and chlorine atoms were bound to the surface as

contaminants from the intermediate step in which SOCl2 was used to convert

the carboxylic group of the 16-PhDA SAMs.

High resolution C 1s (a), O 1s (b), N 1s (c) and F 1s (d) spectra obtained for

the control, 16-PhDA SAM coated and 16-PhDA SAMs functionalised with

Ciprofloxacin® were deconvoluted and are shown in Figure 8-5. The position

of the peak and the contribution of a particular spectra peak depend on the type

of substrate and the chemical deposited on it. From the C 1s spectra, the

control C 1s spectrum was deconvoluted into two components and the peaks at

285.1 eV and 286.6 eV were assigned to C-C and C-O species of the

hydrocarbon [144].


Figure 8-4 XPS survey spectra showing the change in the surface chemistry after

functionalising 16-PhDA SAMs with Ciprofloxacin®. Note the presence of F 1s peak at 687.6

eV and increase in the intensity of N 1s peak at 400.7 eV.

The C 1s spectrum of the 16-PhDA SAM coated Ti6Al4V surface was

deconvoluted into three components. The peaks formed at 285 eV, 286.7 eV

and 289.4 eV were assigned to C-C, C-O and C=O species of the 16-PhDA

SAM molecule [137,156,196]. The deconvoluted C 1s spectrum of SAMs

functionalised with Ciprofloxacin® also rendered three components at 284.8

eV, 286.3 eV and 288.9 eV and are assigned to C-C, C-O and C=O species

[45]. On comparing the C 1s spectra for all three samples, an introduction of

C=O species was observed after SAM coating and drug coating. This carbonyl

(C=O) peak was not observed for the control surface showing that the peak

should essentially be from the SAM and drug molecule. Furthermore, an

increase in the intensity of the C-O peak after functionalising the SAM coated

surface with the drug shows the possibility for the attachment of

Ciprofloxacin® to the 16-PhDA SAMs.

The O 1s peak for the control Ti6Al4V surface was deconvoluted into two

components and the peaks at 530.5 eV and 532.7 eV were assigned to metal

oxide and hydroxyl groups, respectively. On surface modification with 16-

PhDA SAMs, the deconvoluted O 1s spectrum showed three components

including metal oxide (530.3 eV), O-C (531.2 eV) and O=C (532.5 eV) [144].

A small contribution from the P=O at 531.2 eV and P-O at 532.5 eV is also

possible [55]. Also, on functionalisation with Ciprofloxacin®, similar groups

were observed at 530 eV (metal oxides), 531.3 eV (C-O) and 532.6 eV (C=O).


In agreement with C 1s spectra, the O 1s spectra also showed a significant

change confirming the change in surface chemistry.

Figure 8-5 Deconvoluted a) C 1s, b) O 1s, c) N 1s and d) F 1s spectra obtained using XPS for

control, SAM coated and drug coated surfaces.

The deconvoluted N 1s spectrum for the control and SAM coated Ti6Al4V

surfaces showed nitrogen in the form of contaminants (N-C and N-O)

[45,132,181]. On functionalising the SAMs with Ciprofloxacin, the

deconvoluted spectrum showed a significant change. The N-C at 399.8 eV and

amine group (-NH-) at 401.2 eV corresponds to the amino carbons in the

Ciprofloxacin® molecule. High resolution F 1s spectra showed the presence of

fluorine atom only on the Ciprofloxacin functionalised surface and not on the

control and SAM coated surface. These changes in the C 1s, O 1s, N 1s and F

1s confirm the successful functionalisation of 16-PhDA SAMs with

Ciprofloxacin®.

Figure 8-6 shows the C 1s and O 1s spectra obtained for the pure

Ciprofloxacin® powder. Deconvoluted C 1s spectrum showed the presence of


C-C (285 eV), C-O (286.3 eV) and C=O (287.8 eV) [181]. The deconvoluted

O 1s spectra showed O-C and O=C peaks at 531 eV and 532.6 eV

[45,132,181,183]. Also, the Ciprofloxacin® powder showed a single N 1s peak

at 400.7 eV and F 1s peak at 687.5 eV. On comparing these spectra with the C

1s and O 1s spectra obtained for 16-PhDA SAMs functionalised with

Ciprofloxacin®, a good agreement of the peaks were observed. This further

confirmed the successful attachment of Ciprofloxacin® to the 16-PhDA SAMs

coated on SLM fabricated Ti6Al4V surface.

Figure 8-6 Deconvoluted C 1s and O 1s spectra for Ciprofloxacin® powder obtained using

XPS.

The FTIR characterisation results obtained for the drug coated Ti6Al4V

substrates were inconclusive. No significant peak for the expected amide (1640

– 1609 cm-1

or anhydride (1830 – 1800 cm-1

and 1775 – 1740 cm-1

) was

observed in the spectra (see appendix 3). The major factor for this could be the

sample roughness. In FTIR, after the infrared light hits the sample, the

reflected infrared light between the prism and sample provides the information

about the sample’s surface chemistry. In order to get a good signal, a good

contact between the surface of the substrate and the prism is essential.

Although the pressure arm helps ensuring this contact, surface roughness of the

Ti6Al4V substrates could have significantly affected this contact. Also, the

roughness pattern could have affected the beam reflection. As a result of these

issues, the FTIR did not show any significant result to confirm the drug

coating. Hence, analysing rough metal sample surfaces using this pressure arm


supported FTIR may not be the best choice. However, further studies can be

performed by functionalising Ciprofloxacin® to a smooth surface.


In addition to the XPS results, the static contact angle measurements also

confirmed the functionalisation of 16-PhDA SAMs with Ciprofloxacin®. The

transformation of wettable (hydrophilic) surface observed before surface

modification to a non-wettable surface (hydrophobic) after drug attachment

shows a change in the surface wettability (Figure 8-7). This hydrophobic

nature can be attributed to the presence of hydrocarbon in the terminal group of

Ciprofloxacin® molecules. The ODPA SAMs functionalised with

Ciprofloxacin® did not show a significant difference since the surface was

already hydrophobic. Also, as stated earlier, due to the non-availability of a

functional group in the terminal end, the ODPA SAMs cannot be

functionalised with Ciprofloxacin®.

Figure 8-7 Static contact angle measurements using a contact angle goniometer. a) Before

cleaning; b) after cleaning; c) after 16-PhDA SAM attachment and d) after immobilisation of

16-PhDA SAMs with Ciprofloxacin®.


8.4. Stability Studies

8.4.1. Oxidative stability of Ciprofloxacin®

8.4.1.1. Surface chemistry

Figure 8-8 gives detailed high resolution XPS spectra of the convoluted peaks

for the C 1s, O 1s, N 1s and F 1s regions under oxidative conditions. The peaks

for all elements were observed to be very consistent for the whole duration of

the oxidative exposure.

Figure 8-8 Relative intensity of C 1s, O 1s, N 1s and F 1s spectra obtained using XPS for

samples exposed to oxidative conditions for different time intervals.


Table 8-2 XPS determined relative atomic percentages of elements detected for different

oxidative exposure time intervals. The calibration error during peak integration in XPS is ±

10%.

8.4.1.2. Surface wettability

The wettability of the Ciprofloxacin® coated surfaces was measured to

determine the stability of the drug under oxidative conditions. The surfaces

remained hydrophobic during all oxidative exposure time intervals and their

Detected Elements Relative Atomic Percentage (%)

After

Immobilis

ation

7 Days 14 Days 28 Days 42 Days

Al 2p

Mean 1.24 1.06 0.86 0.96 1.01

S.E ± 0.1 ± 0.2 ± 0.2 ± 0.2 ± 0.1

Ti 2p

Mean 4.68 3.9 4.29 3.87 3.7

S.E ± 0.3 ± 0.5 ± 0.4 ± 0.3 ± 0.2

C 1s

Mean 60.59 62.55 61.75 61.12 63.03

S.E ± 1.1 ± 1 ± 0.6 ± 0.7 ± 1.5

F 1s

Mean 1.5 1.58 1.28 1.2 1.34

S.E ± 0.1 ± 0.1 ± 0.1 ± 0.2 ± 0.1

N 1s

Mean 4.14 3.5 4.07 4.22 3.46

S.E ± 0.3 ± 0.3 ± 0.1 ± 0.3 ± 0.5

O 1s

Mean 24.84 24.4 24.63 25.57 24.87

S.E ± 0.5 ± 0.6 ± 1 ± 0.4 ± 1

P 2p

Mean 2.23 2.1 2.39 2.03 1.94

S.E ± 0.3 ± 0.1 ± 0.2 ± 0.1 ± 0.1

S 2p

Mean 0.37 0.39 0.32 0.61 0.31

S.E ± 0.1 ± 0.1 ± 0.1 ± 0.1 ± 0.1

Cl 2p

Mean 0.41 0.44 0.41 0.44 0.33

S.E ± 0.1 ± 0.1 ± 0.1 ± 0.1 ± 0.1


respective contact angles are shown in Figure 8-9. As explained earlier, the

hydrophobic nature of the surface is due to the presence of hydrocarbons (from

the drug) on the outer most surface. Surface wettability measurements showed

that the drug is highly stable and remain attached to the 16-PhDA SAM coated

Ti6Al4V surface under oxidative (ambient air) conditions. Bhure et al. [55]

also observed a small change in the contact angle (~ 10°) within the

hydrophobicity limit during their oxidative stability measurement. Although a

small change was observed in the present study, this was not significant.

Hence, this confirms that the attached Ciprofloxacin® is stable under oxidative

condition used in this study.

Figure 8-9 Static water contact angles obtained to determine the stability of Ciprofloxacin®

under oxidative conditions at different time intervals.

8.4.2. In vitro stability of Ciprofloxacin®

8.4.2.1. Surface chemistry

The reaction mechanism for the release of Ciprofloxacin® under in vitro

condition is hydrolysis. As a result, the immobilised drug will release from the

SAM molecule and regain its structure to produce therapeutic effect. Table 8-3

shows a change in the XPS determined relative atomic percentage of elements

after soaking the Ciprofloxacin® coated surface in Tris-HCl buffer solution at

different time intervals. It can be observed from the table that the concentration

of C and F decreased with the increase in the immersion time period.

93 97 98

106

90

0

20

40

60

80

100

120

Week 0 Week 1 Week 2 Week 4 Week 6

Co

nta

ct

An

gle

(°)


Concurrently, the elemental concentration of Ti, O and Al increased. This

suggests the hydrolysis of drug from the terminal end of 16-PhDA SAMs.

Table 8-3 XPS determined relative atomic percentages of elements detected for different

immersion time interval in Tris-HCl buffer solution. The calibration error due to XPS peak

integration is ± 10%.

Detected

Elements

Relative Atomic Percentage (%)

Drug Immobilisation 7 Days 14 Days 28 Days 42 Days

Before After

Al 2p

Mean 2.39 1.24 1.16 1.57 1.5 2

S.E ± 0.2 ± 0.2 ± 0.2 ± 0.1 ± 0.2 ± 0.1

Ti 2p

Mean 7.57 4.68 5.54 5.88 6.44 6.88

S.E ± 0.4 ± 0.2 ± 0.1 ± 0.2 ± 0.8 ± 0.6

C 1s

Mean 53.68 60.59 58.17 56.55 56.7 55.12

S.E ± 0.5 ± 1.1 ± 0.1 ± 0.9 ± 1.6 ± 0.3

F 1s

Mean 0 1.5 1.18 1.15 1.16 0.92

S.E 0 ± 0.1 ± 0.1 ± 0.1 ± 0.1 ± 0.1

N 1s

Mean 0.82 4.14 3.84 4.05 3.97 4.01

S.E ± 0.1 ± 0.2 ± 0.1 ± 0.4 ± 0.3 ± 0.3

O 1s

Mean 31.05 24.84 27.85 27.99 27.5 28.46

S.E ± 0.4 ± 0.5 ± 0.2 ± 0.6 ± 1 ± 0.3

P 2p

Mean 3.51 2.23 2 2.34 1.9 1.76

S.E ± 0.2 ± 0.3 ± 0.1 ± 0.2 ± 0.2 ± 0.1

Si 2p

Mean 1.98 0 0.14 0.35 0.6 0.6

S.E ± 0.6 0 ± 0.1 ± 0.1 ± 0.2 ± 0.1

S 2p

Mean 0 0.37 0 0 0 0

S.E 0 ± 0.1 0 0 0 0

Cl 2p

Mean 0 0.41 0 0 0 0

S.E 0 ± 0.1 0 0 0 0

Ba 3d

Mean 0 0 0.11 0.14 0.25 0.25

S.E 0 0 ± 0.1 ± 0.1 ± 0.1 ± 0.1

Figure 8-10 shows the high resolution spectra of C 1s, O 1s, N 1s and F 1s

region obtained after different immersion time intervals. From the C 1s region,

reduction in the intensity of C-C, C-O and C=O peaks with an increase in

immersion time was observed; whereas, the O 1s region showed an increase in

the intensity of the metal oxide peak at 530 eV and a decrease in the O-C peak.

The N 1s region showed a change in the intensity but a clear trend was not


observed. This may be due to the adsorption of nitrogenous contaminants from

air and Tris-HCl buffer. The relative intensity of the fluorocarbon peak

obtained at 687.5 ± 0.2 eV was observed to decrease with the increase in

immersion time and clearly shows the desorption of Ciprofloxacin® from the

SAM coated surface. It can be observed from the F 1s spectra that nearly half

of the drug was desorbed after 28 days with some drug remaining after 42

days. A small amount of barium was observed only for the samples soaked in

buffer solution. This barium may be contamination from the glass vial used to

immerse the samples in Tris-HCl buffer solution.

Figure 8-10 High resolution XPS spectra of C 1s, O 1s, N 1s and F 1s regions showing their

relative intensities after the immersion of Ciprofloxacin® coated surfaces for different time

intervals.

8.4.2.2. Surface wettability

Under in vitro conditions, the stability of Ciprofloxacin® on the 16-PhDA

SAM coated Ti6Al4V surface was significantly different to its stability under


oxidative condition. After incubating the drug coated surfaces in Tris-HCl

buffer solution, the contact angle values changed with respect to its initial

value before immersion (Figure 8-11). This change can be attributed to the

cleaving of Ciprofloxacin® molecules from the tail group of SAMs to the Tris-

HCl buffer solution. On desorption of Ciprofloxacin® from the surface, the

surface will be expected to be hydrophilic. This is due to the fact that a bare

Ti6Al4V metal surfaces and 16-PhDA SAM coated metal surfaces are

hydrophilic.

Figure 8-11 Static water contact angle exhibiting a change in the surface wettability after the

immersion of Ciprofloxacin® coated surface in Tris-HCl buffer solution for different time

intervals.

In the present study, after the immersion of the drug coated surfaces in Tris-

HCl buffer solution, the contact angles were observed to decrease compared to

their initial values after drug coating (Figure 8-11). The wettability

measurement showed desorption of drugs from the surface with a clear trend.

The contact angles decreased with the increase in immersion time revealing

desorption of Ciprofloxacin®. However, the calculated standard deviations

were high. This is likely to be due to the presence of Ciprofloxacin® in a few

islands and not on the other. This can be due to an irregular assembly and

cleaving of monolayers after they are exposed to in vitro solution.

93

73 67

43

34

0

20

40

60

80

100

120

Week 0 Week 1 Week 2 Week 4 Week 6

Co

nta

ct

An

gle

(°)


8.5. Drug Quantification

The total amount of drug loaded was calculated using UV-Vis

spectrophotometry. A calibration curve was plotted for Ciprofloxacin® in 10

mM NaOH (pH ~ 7.4) for the concentration range from 1 ng/mL to 1 µg/mL.

The regression equation was determined as y =2.34x X 10-4

+ 2.4 X 10-3

(R2

=

0.996). The total amount of Ciprofloxacin® immobilised to SAMs was

determined to be from 1.17 ± 0.1 µg/cm2 (assuming the surface area of the

sample is 3.2 cm2). Previous literature showed the immobilisation of ~100

ng/cm2 of Paclitaxel™ to 16-PhDA SAMs [144]. However, in the case of this

study a drug loading ten times greater was obtained. This variation is more

likely due to the surface roughness pattern of the SLM fabricated sample.

Although the top surface was mechanically polished, the other faces remained

rough. The change in the surface area can influence the amount of SAM

adsorption which will then directly influence the total amount of drug loaded.

Also, the Paclitaxel™ (853.91 g/mole) drug has a molecular mass more than

twice that of the Ciprofloxacin® (331.35 g/mole). Ajami et al. [136] reported

the attachment of Ciprofloxacin® to SAMs, but the amount of drug coated was

not quantified. However, this study shows that by increasing the surface area,

more drugs can be immobilised to 16-PhDA SAMs.

8.6. Drug Elution Studies

Figure 8-12 shows the percentage of drug eluted to the Tris-HCl buffer

solution at 7, 14, 28 and 42 day time intervals. The calibration curve was

obtained for Ciprofloxacin® in Tris-HCl buffer over the concentration range

from 1 ng/mL to 1 µg/mL and the regression equation for Ciprofloxacin® in

Tris-HCl buffer (pH ~ 7.4) was determined as y = 10x X 10-5

+ 1.2 X 10-3

(R2

= 0.998). UV-Vis determined drug release profile showed a sustained release

profile. Previously reported studies have shown the possibility for the surface

to have both covalently bound drug and a reservoir of drug as the surface is

rough [136,144]. In the present study, such formation of reservoir of

Ciprofloxacin® was not witnessed. This may be because the samples after the

immobilisation of Ciprofloxacin® were sonicated with anhydrous

tetrahydrofuran and deionised water twice for two minutes. During this


procedure, it can be expected that most of the physisorbed drug molecule

would be removed. Also the amount of drug used for immobilisation is less.

Despite these reasons, there is the possibility for some physisorbed drug

molecule to be present. If there was a reservoir of drug, a bulk release of the

drug should have been witnessed during the UV-Vis spectroscopic

measurement. However, the 42 days in vitro studies revealed no such bulk

release of the drug. This shows that the presence of a reservoir of

Ciprofloxacin® on the samples is less likely. Thus, as there was no initial burst

release, it can be stated that most of the drug molecules were covalently bound.

Figure 8-12 Percentage drug released at different time intervals determined using UV-Visual

Spectrophotometer.

After immersing the sample in the buffer solution for 7 days, nearly 13% of the

drug eluted whereas after a 28 day interval, nearly 50% of the drug had been

released. However, after 42 days of immersion, nearly 40% of the drug was

still observed to be eluted from the surface. The UV-Vis determined release

profile was observed to be in good agreement with the XPS determined in vitro

stability. This may suit biomedical applications where a sustained release of

the drug is required over a period of time. However, it should be noted that

Ciprofloxacin® will be inactive when it is immobilised to the monolayers and

will only be active upon its release from the monolayer. The release of the drug

in its active form from the monolayer can be attained by hydrolysis.

0

20

40

60

80

100

0 Day 7 Days 14 Days 28 Days 42 Days

To

tal am

ou

nt

of

dru

g r

ele

ased

in

p

erc

en

tag

e (

%)

Sample immersion time intervals

Percentage drug release

Sustained drug release


Ajami et al. [136] reported the release of Ciprofloxacin® from a silane SAMs.

In their study, more than 50% of the drug was observed to release within 5

days in 20 mL of high purity water (18.2MΩ.cm) maintained at 37 °C. They

have also observed 80% release in 26 days. There is a high possibility for this

release profile to be faster in physiological condition than in water. Release

profile reported by Mani et al. [37] for flufenamic acid showed an initial burst

release followed by a sustained release of the drug. In an another study

reported by Mani et al. [144], they observed a burst release of Paclitaxel™ in 7

days followed by a sustained release for up to 35 days. Comparing the release

profiles reported by these researchers with this study, the drug release is slow

and exhibits the potential to use this system for long-term release (more than 6

weeks) of drug from an implant’s surface. As mentioned before, the slow

release is mainly due to the covalent bonding of the drug and non-availability

of physisorbed drug on the surface.

This release profile further confirms the formation of an anhydride rather than

amide and shows possibility for the attachment of Ciprofloxacin® to the acyl

chloride via a carboxylic group. This is because amide is a poor leaving group

with a pKa of about 25. Hence, hydrolysis of amides requires vigorous reaction

conditions (such as addition of catalysts and high temperature ~ 100 °C) [172].

In this study, the drug coated implants were immersed in Tris-HCl buffer

solution of pH 7.4 at 37.5 °C. This pH is slightly towards the basic condition

and the temperature used is relatively low. Under this mild condition,

hydrolysis of amides is less favourable, whereas, anhydrides will readily

hydrolyse (since pKa of anhydrides is about 5) in water to form corresponding

carboxylic acid [172]. Thus the formation of anhydrides rather than amides is

most expected.

8.7. Antibacterial Susceptibility

8.7.1. Test-1

The susceptibility test was performed to determine if the immobilised drug was

releasing and to investigate activity of the drug after hydrolysing from the

monolayers. Figure 8-13 shows the photographic images of the zone of


inhibition for the control discs and discs coated with Ciprofloxacin®. Release

of the drug from the standard filter paper disc coated with the drug was

observed to inhibit the growth of S. aureus whereas the metal disc coated with

drug did not show any drug elution and inhibition. In contrast, the metal discs

coated with Ciprofloxacin® were observed to inhibit the growth of E.coli. A

similar trend was followed in all five sets for both bacteria. It may be argued

that the metal discs placed on S. aureus were not coated with the drug, but this

may not be the case. All metal discs used for both S. aureus and E. coli were

from the same batch. Also, the XPS studies on the drug coated metal discs

confirmed the functionalisation of Ciprofloxacin® (Figure 8-14).

Figure 8-13 Antibacterial susceptibility test against S. aureus (a) and E. coli (b). Label

description: A - control standard filter paper disc without drug, B - standard filter paper disc

with drug; C - Control Ti6Al4V metal disc without drug; D - metal disc coated with drug.

Ciprofloxacin®, being a wide spectrum antibiotic, should inhibit the growth of

S. aureus and previous studies have proved this [175]. Whereas, the results

obtained in this study for S. aureus are unclear. One possible reason for not

inhibiting the growth may be due to the temperature used to incubate the

culture plates after placing the discs. Since the S. aureus inoculated plates were

maintained at 30 °C, this might have affected the hydrolysis of the drug from

the monolayer. Since the drugs on the standard filter paper discs were

physisorbed, it has released the drug and inhibited the growth. The temperature

used to incubate E. coli (37.5 °C) was observed to hydrolyse the drug from the

metal disc and inhibit the bacterial growth. Thus the temperature might have

been the factor that affected the drug release.

After the experimentation time, the samples were removed from the culture

plate and rinsed with ethanol for 2 minutes before examining the presence of


Ciprofloxacin® using XPS on these discs. Figure 8-14 shows the XPS spectra

obtained for a control, drug coated, drug coated metal disc from S. aureus

culture and drug coated metal disc from on E. coli culture.

Figure 8-14 XPS spectra of F 1s region for control, drug coated and drug coated discs placed

on S. aureus and E. coli bacteria.

The spectra clearly show the existence of the drug on the surface of the

Ti6Al4V discs placed on the culture medium; however, their intensities varied.

The variation in the intensities might be due to the fact that after placing on the

medium, adherence of the agar medium to the metal discs are possible. Since

XPS probes typically 1-10 nm of the surface, the adhered medium might have

reduced the penetration depth of the photoelectrons. Although the results

showed the drug released from the metal substrate is active for E. coli, no

activity was observed for S. aureus and the reason for this should be further

explored. In addition to the drug release profile, the release of drug in this

bacterial culture (again a mild condition) further shows the possibility for the

formation of anhydrides rather than amide during the reaction between the acyl

chloride and Ciprofloxacin®.

682 684 686 688 690 692 694 696

Rela

tiv

e I

nte

nsit

y (

Arb

itra

ry U

nit

s)

Binding Energy (eV)

Control (MC)

Drug coated

After placement on the S. aureus culture

After the placement on the E. coli culture


8.7.2. Test-2

Since the drug eluted from the metal discs did not show any activity against the

gram positive bacteria S. aureus, this test was mainly performed to investigate

antibacterial susceptibility of Ciprofloxacin® released from the metal samples

in Tris-HCl buffer. This test was also performed to qualitatively observe if

there was an increase in the inhibition zone diameter with respect to immersion

time intervals in the Tris-HCl buffer solution. Theoretically, as the immersion

time of the drug coated sample in Tris-HCl is increased, the amount of drug

eluted from the metal surface should increase. As the concentration of the drug

in Tris-HCl buffer increases, there should be an increase in the diameter of the

inhibition zone. Figure 8-15 shows the photographic images of the zone of

inhibition for the discs coated with Ciprofloxacin® eluted at different time

intervals.

Figure 8-15 Antibacterial susceptibility test against a) S. aureus and b) E. coli. Label

description: C- control disc with no drug; CD – control with drug on; TC: discs coated with

Tris-HCl buffer; 1, 2, 4, 6 – immersion time intervals (in weeks) of the samples in Tris-HCl

buffer solution.

It can be observed from the graph that the discs coated with the drug showed

anti-bacterial activity confirming the drug was active upon their release.

Furthermore, the discs coated with drugs eluted at different time intervals

showed an increased zone of inhibition diameter (Figure 8-16) for both S.

aureus and E.coli. As expected this may be due to the fact that as the sample

immersion time was increased, more Ciprofloxacin® molecules eluted from

the surface to the buffer solution thereby increasing the drug concentration.

This study qualitatively showed the release of the drug and its antibacterial


activity. Since the present study did not use standard discs coated with

Ciprofloxacin®, the MIC for S. aureus and E. coli was not determined.

Figure 8-16 Inhibited zone diameters for S. aureus and E. coli.

8.8. Summary

The use of 16-PhDA monolayers to modify SLM fabricated structures to

deliver therapeutics has been demonstrated. Experimental results showed a

covalent attachment of the Ciprofloxacin® molecules to SAMs and sustained

release in Tris-HCl buffer solution. The total drug immobilised to the

monolayers was estimated to be approximately 1.2 ± 0.1 µg/cm2. The

antibacterial susceptibility study revealed that the drug was active upon its

release. Thus the attachment, stability and the delivery of Ciprofloxacin® from

SLM fabricated surface shows the potential to use this approach to deliver

therapeutics directly from customised implant surfaces.

0

5

10

15

20

Control Controlwith Tris-HCl buffer

Controlwith drug

Week 1 Week 2 Week 4 Week 6

Inh

ibit

ed

zo

ne w

idth

(m

m)

Gram positive

Gram Negative

Chapter 9: Conclusion and Future Work 158

9. CONCLUSION AND FUTURE WORK

9.1. Conclusion

This research concludes the potential to integrate SLM with SAM based

surface modification to deliver therapeutics directly from customised implant

surfaces fabricated by SLM. Based on the experimental studies, this research

has also arrived at the following conclusions.

The surface chemistry of the SLM fabricated parts was not

homogeneous and varied depending on their exposure to the laser

beam. Aluminium rich areas were witnessed on the SLM fabricated

Ti6Al4V parts when a high laser energy density was applied. The

surface chemistries of the SLM-AF and SLM-MP were different with a

varied elemental contribution.

16-PhDA SAMs were covalently bound to both the SLM-AF and SLM-

MP Ti6Al4V surfaces and were stable for 28 days on these surfaces.

The surface roughness of the SLM-AF did not have a significant

impact on the in vitro stability of the 16-PhDA SAMs.

The use of acid chloride esterification was successful to immobilise the

Paracetamol to the 16-PhDA SAMs. However, the use of Paracetamol

may not be the best choice to prove the accomplishment of the reaction

due to the possibility that the results may be affected by atmospheric

contaminants. For example, a drug molecule with distinct elements

such as fluorine and bromine that are unlikely to be present in the

atmosphere can prevent contamination affecting the results. Also,

Paracetamol (phenolic esters) can easily hydrolyse from the SAM

molecules in the presence of moisture or water molecules. Thus, a

careful consideration is required when choosing a model drug.

The total amount of Ciprofloxacin® coated on the SLM-MP Ti6Al4V

was estimated to be approximately 1.2 µg/cm2. The drug was observed

to be highly stable under oxidative conditions for over 42 days. The in

vitro release of the drug in Tris-HCl buffer solution showed the

covalent attachment of the drug to the monolayers and exhibited the


potential to use this model for applications requiring a sustained release

pattern. Upon release from the monolayer, the drug was active and

inhibited the growth of bacteria.

Contamination of Ti6Al4V powder due to atmospheric contaminants

including carbon and nitrogen was witnessed upon recycling the

powders in a SLM machine.

9.2. Future Work

This work, being the first study of its kind in AM, has opened up future studies

in various directions. The following are some recommendations to extend the

knowledge beyond that performed in this work.

This study compared the surface chemistry of Ti6Al4V surface using

XPS only. These results can be compared with other surface chemical

characterisation techniques such as AES for improved understanding.

Estimating the oxide layer thickness and surface chemistry of the SLM

as-fabricated surface in this study was limited due to the sensitivity of

XPS to rough surfaces. However, there are other SLM machines such

as EOSINT M 280, EOS M 400 which provide a better surface finish.

Studies on the surface chemistry of these parts could be used for

comparison.

Electron beam melting (EBM), an AM technique, is also used for the

fabrication of biomedical devices. Comparison of the surface

chemistries obtained by EBM and SLM would be beneficial.

This study showed the attachment and functionalisation of 16-PhDA

monolayers only. Other phosphonic acid monolayers could be used and

their stability could be analysed and compared.

Although the surface modification and functionalisation were

performed by immersing the whole SLM fabricated part into the SAM

and drug solution, only the top surface was studied and characterised.

Implants often have complex geometries and not just a flat surface.

Hence, a study on the immobilisation of drugs to SAMs adsorbed on

the whole of the surface is required.


This study proposes the reaction of acyl chloride with carboxylic acid

to form an anhydride but not the reaction of acyl chloride with amine to

form an amide. This has to be further confirmed by the use of FTIR-

ATR to know the exact reaction mechanism.

The amount of drug functionalised to SAMs in this study may be low

for certain applications and hence methods need to be developed to

increase the amount of drug loading, for example, growing multi-layers

by using polymer brushes can be explored.

Although the sustained release pattern of the drug is convincing, a

shorter time gap between the intervals over the study period will

generate improved knowledge on the drug release profile.

Metal discs coated with Ciprofloxacin® and incubated at 30°C for

evaluation of its antibacterial susceptibility did not show drug release.

Although temperature might have been a factor, this should be further

explored.

This study proposed the attachment of only therapeutic drugs to SAMs.

The SAM attached SLM fabricated Ti6Al4V surfaces can be explored

for the immobilisation of proteins which would be useful for various

biomedical applications.

A similar procedure can be used to functionalise other metallic

biomaterial surfaces (including 316L SS and Co-Cr) and their stability

can be tested.

Often, for biomedical applications, more than one type of drug is

required and/or delivery of drug from a particular area is preferred.

Hence, patterning of SLM fabricated surfaces with SAMs and

functionalising the SAMs with various drugs would be beneficial.

References 161

REFERENCES

1. Ratner BD. A history of biomaterials. In: Ratner BD, Hoffman A s,

Schoen FJ, Lemons JE, editors. Biomater. Sci. an Introd. to Mater. Med.

second edi. London: Elsevier; 2004. p. 10–23.

2. Grillo HC. Diseases and results of treatment. In: Grillo HC, editor. Surg.

trachea bronchi. London: BC Decker Inc, USA; 2004. p. 173–429.

3. Manivasagam G, Dhinasekaran D, Rajamanickam A. Biomedical

implants : corrosion and its prevention - a review. Recent Patents

Corros. Sci. 2010;2:40–54.

4. Paital SR, Dahotre NB. Calcium phosphate coatings for bio-implant

applications: Materials, performance factors, and methodologies. Mater.

Sci. Eng. R Reports. 2009;66(1-3):1–70.

5. Hassler C, Boretius T, Stieglitz T. Polymers for neural implants. J.

Polym. Sci. Part B Polym. Phys. 2011 Jan 1;49(1):18–33.

6. 10th Annual Report. National Registry for England, Wales and Northern

Ireland [Internet]. 2013. Available from:

http://www.njrcentre.org.uk/njrcentre/Portals/0/Documents/England/Re

ports/10th_annual_report/NJR 10th Annual Report 2013 B.pdf

7. Miller RS. Increasing incidence of joint replacements burdens

healthcare system. Med. Device Marketpl. Rev. 2008 p. 1–3.

8. Paxton EW, Inacio M, Slipchenko T, Fithian DC. The kaiser

permanente national total joint replacement registry. Perm. J.

2008;12(3):12–6.

9. King M. The future of the orthopedic devices market to 2012 [Internet].

2008. Available from:

http://www.hipresurfacingnews.com/archives/265-The-Future-of-the-

Orthopedic-Devices-Market-to-2012.html. Accessed on 30th

January,

2012.

10. Singhatanadgit W. Biological responses to new advanced surface

modifications of endosseous medical implants. Bone Tissue Regen.

Insights. 2009;2:1–11.

11. Anderson JM. Biological responses to materials. Annu. Rev. Mater.

Res. 2001;31:81–110.

12. Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to

biomaterials. Semin. Immunol. 2008;20(2):86–100.

References 162

13. Castner DG, Ratner BD. Biomedical surface science: foundations to

frontiers. Surf. Sci. 2002;500(1-3):28–60.

14. Eyers D, Dotchev K. Technology review for mass customisation using

rapid manufacturing. Assem. Autom. 2010;30(1):39–46.

15. Hao L, Harris R. Customised implants for bone replacement and

growth. In: Bartolo P, Bidanda B, editors. Bio-materials Prototyp. Appl.

Med. Springer US; 2008. p. 79–107.

16. Taylor M, Tanner KE. Fatigue failure of cancellous bone: a possible

cause of implant migration and loosening. J. bone Jt. surgery. 1997;79-

B(2):181–2.

17. Freitag L. Airway stents. Eur Respir Mon. 2010;48:190–217.

18. Huddleston J, Gobezie R, Rubash H. Custom Implants in Revision Total

Knee Arthroplasty. In: Bono J V, Scott RD, editors. Revis. Total Knee

Arthroplast. Springer US; 2005. p. 105–16.

19. Freeman R, Plante-bordeneuve P. Early migration and late aseptic

failure of proximal femoral prostheses. J. bone Jt. surgery.

1994;76:432–8.

20. Wholey MH, Ferral H, Reyes R, Lopera J, Castan W. Retrieval of

migrated colonic stents from the rectum. Cardiovasc. Intervent. Radiol.

1997;20:477–80.

21. Galindo-Moreno P, Padial-Molina M, Sánchez-Fernández E,

Hernández-Cortés P, Wang H-L, O'Valle F. Dental implant migration in

grafted maxillary sinus. Implant Dent. 2011;20(6):400–5.

22. Hebel KS, Gajjar RC. Cement-retained versus screw-retained implant

restorations: achieving optimal occlusion and esthetics in implant

dentistry. J. Prosthet. Dent. 1997;77(1):28–35.

23. Shadid R, Sadaqa N. A comparison between screw- and cement-retained

implant prostheses. A literature review. J. Oral Implantol.

2012;38(3):298–307.

24. Chaar MS, Att W, Strub JR. Prosthetic outcome of cement-retained

implant-supported fixed dental restorations: a systematic review. J. Oral

Rehabil. 2011;38(9):697–711.

25. Natali AN, Carniel EL, Pavan PG. Dental implants press fit phenomena:

biomechanical analysis considering bone inelastic response. Dent.

Mater. 2009;25(5):573–81.

26. Lu J. Orthopaedic bone cements. In: Poitout DG, editor. Biomech.

Biomater. Orthop. First. Springer New York; 2004. p. 96.

References 163

27. Bibb R, Eggbeer D, Evans P. Rapid prototyping technologies in soft

tissue facial prosthetics: current state of the art. Rapid Prototyp. J.

2010;16(2):130–7.

28. Wood D. Tracheal and Bronchial Stenting. In: Grillo H, editor. Surg.

Trachea Bronchi. Ontario: BC Decker Inc, Hamilton, Ontario; 2004. p.

763–90.

29. Stoeckel D, Pelton A, Duerig T. Self-expanding nitinol stents: material

and design considerations. Eur. Radiol. 2004;14(2):292–301.

30. Kurella A, Dahotre NB. Review paper: surface modification for

bioimplants: the role of laser surface engineering. J. Biomater. Appl.

2005;20(1):5–50.

31. Aksakal B, Yildirim O., Gul H. Metallurgical failure analysis of various

implant materials used in orthopedic applications. J. Fail. Anal. Prev.

2004;4(3):17–23.

32. Bhola R, Bhola SM, Mishra B, Olson DL. Corrosion in titanium dental

implants / prostheses - a review. Trends Biomater Artif Organs.

2011;25(1):34–46.

33. Pye D, Lockhart DE, Dawson MP, Murray C, Smith J. A review of

dental implants and infection. J. Hosp. Infect. 2009;72(2):104–10.

34. Knobben BS, van Horn JR, van der Mei HC, Busscher HJ. Evaluation of

measures to decrease intra-operative bacterial contamination in

orthopaedic implant surgery. J. Hosp. Infect. 2006;62(2):174–80.

35. Leung JW. History of bile duct stenting: rigid prostheses. In: Kozarek R,

Baron T, Song H-Y, editors. Self-expandable Stents Gastrointest. Tract.

New York, NY: Springer New York; 2013. p. 15–31.

36. Noppen M, Piérard D, Meysman M, Claes I, Vincken W. Bacterial

colonization of central airways after stenting. Am. J. Respir. Crit. Care

Med. 1999;160(2):672–7.

37. Mani G, Johnson DM, Marton D, Feldman MD, Patel D, Ayon A and

Agarwal M. Drug delivery from gold and titanium surfaces using self-

assembled monolayers. Biomaterials. 2008;29(34):4561–73.

38. An YH, Dickinson RB, Doyle RJ. Mechanisms of bacterial adhesion

and pathogenesis of implant and tissue infections. In: An YH, Friedman

RJ, editors. Handb. Bact. Adhes. Princ. Methods Appl. New Jersy:

Humana Press Inc; 2002. p. 1–27.

39. Harris LG, Richards RG. Staphylococci and implant surfaces: a review.

Injury. 2006;37 Suppl 2:S3–14.

References 164

40. Garvin K, Feschuk C. Polylactide-polyglycolide antibiotic implants.

Clin. Orthop. Related. Res. 2005;437:105–10.

41. Schmidmaier R, Simsek M, Baumann M, Emmerich P, Meinhardt G.

Synergistic antimyeloma effects of zoledronate and simvastatin.

Anticancer Drugs. 2006;17(6):621–30.

42. Stone RM, O’Donnell MR, Sekeres MA. Acute myeloid leukemia. Am.

Soc. Hematol. 2004;98–117.

43. Levy Y, Mandler D, Weinberger J, Domb AJ. Evaluation of drug-

eluting stents’ coating durability - clinical and regulatory implications.

J. Biomed. Mater. Res. B. Appl. Biomater. 2009;91(1):441–51.

44. Otsuka Y, Chronos N, Apkarian R, Robinson K. Scanning electron

microscopic analysis of defects in polymer coatings of three

commercially available stents: comparison of BiodivYsio, Taxus and

Cypher stents. J. Invasive Cardiol. 2007;19(2):71–6.

45. Mani G, Macias CE, Feldman MD, Marton D, Oh S, Mauli Agrawal C.

Delivery of paclitaxel from cobalt-chromium alloy surfaces without

polymeric carriers. Biomaterials. 2010;31(20):5372–84.

46. Knetsch MLW, Koole LH. New strategies in the development of

antimicrobial coatings: The example of increasing usage of silver and

silver nanoparticles. Polymers (Basel). 2011 Jan 26;3(4):340–66.

47. Dubey M, Weidner T, Gamble LJ, Castner DG. Structure and order of

phosphonic acid-based self-assembled monolayers on Si(100).

Langmuir. 2010;26(18):14747–54.

48. Faucheux N, Schweiss R, Lützow K, Werner C, Groth T. Self-

assembled monolayers with different terminating groups as model

substrates for cell adhesion studies. Biomaterials. 2004;25(14):2721–30.

49. Arabnejad Khanoki S, Pasini D. Multiscale design and multiobjective

optimization of orthopedic hip implants with functionally graded

cellular material. J. Biomech. Eng. 2012;134(3):031004–10.

50. Chahine G, Atharifar H, Smith P, Kovacevic R. Design optimization of

a customized dental implant manufactured via Electron Beam Melting ®

Int. Solid Free. Fabr. Symp. 2009. p. 631–40.

51. Kruth J, Badrossamay M, Yasa E, Deckers J, Thijs L, Humbeeck J Van.

Part and material properties in selective laser melting of metals. 16th

International Symposium on Electromachining. 2010. p. 1–12.

52. Mani G, Feldman MD, Patel D, Agrawal CM. Coronary stents: a

materials perspective. Biomaterials. 2007;28(9):1689–710.

References 165

53. Mani G, Johnson DM, Marton D, Dougherty VL, Feldman MD, Patel D,

and Ayon A. Stability of self-assembled monolayers on titanium and

gold. Langmuir. 2008;24(13):6774–84.

54. Torres N, Oh S, Appleford M, Dean DD, Jorgensen JH, Ong JL,

Agarwal MC and Mani G. Stability of antibacterial self-assembled

monolayers on hydroxyapatite. Acta Biomater. 2010;6(8):3242–55.

55. Bhure R, Abdel-Fattah TM, Bonner C, Hall F, Mahapatro A. Stability of

phosphonic self assembled monolayers (SAMs) on cobalt chromium

(Co-Cr) alloy under oxidative conditions. Appl. Surf. Sci.

2011;257(13):5605–12.

56. Kaufmann C, Mani G, Marton D, Johnson D, Agrawal CM. Long-term

stability of self-assembled monolayers on electropolished L605 cobalt

chromium alloy for stent applications. J. Biomed. Mater. Res-A.

2011;98B(2):280–9.

57. Tegoulia V, Cooper S. Staphylococcus aureus adhesion to self-

assembled monolayers: effect of surface chemistry and fibrinogen

presence. Colloids Surfaces B Biointerfaces. 2002;24(3-4):217–28.

58. Mahapatro A, Johnson DM, Patel DN, Feldman MD, Ayon A, Agrawal

CM. The use of alkanethiol self-assembled monolayers on 316L

stainless steel for coronary artery stent nanomedicine applications: an

oxidative and in vitro stability study. Nanomedicine. 2006;2(3):182–90.

59. Schwartz DK. Mechanisms and kinetics of self -assembled monolayer

formation. Annu. Rev. Phys. Chem. 2001;52:107–37.

60. Lindner M, Hoeges S, Meiners W, Wissenbach K, Smeets R, Telle R,

Poprawe Reinhart and Fischer Horst. Manufacturing of individual

biodegradable bone substitute implants using selective laser melting

technique. J. Biomed. Mater. Res-A. 2011;97(4):466–71.

61. Yadroitsev I, Smurov I. Surface morphology in selective laser melting

of metal powders. Phys. Procedia. 2011;12:264–70.

62. ASTM-F2792. Standard Terminology for Additive Manufacturing

Technologies. Standard Terminologies for Additive Manufacturing.

Technology, ASTM International, West Conshohocken, PA, 2003;

2010.

63. Gibson I, Rosen DW, Stucker B. Additive Manufacturing Technologies.

Boston, MA: Springer US; 2010.

64. Santos E, Shiomi M, Osakada K, Laoui T. Rapid manufacturing of

metal components by laser forming. Int. J. Mach. Tools Manuf.

2006;46(12-13):1459–68.

References 166

65. Schleifenbaum H, Meiners W, Wissenbach K, Hinke C. Individualized

production by means of high power Selective Laser Melting. CIRP J.

Manuf. Sci. Technol. CIRP; 2010;2(3):161–9.

66. Tuck C, Hague R. The pivotal role of rapid manufacturing in the

production of cost effective customised products. Int. J. Mass Cust.

2006;1(3):360–73.

67. Hopkinson N, Hague RJ, Dickens PM. Rapid Manufacturing: An

Industrial Revolution for the Digital Age. 1st ed. Hopkinson N, Hague

RJ, Dickens PM, editors. John Wiley and Sons Ltd., West Sussex, UK.;

2006.

68. Li R, Shi Y, Wang Z, Wang L, Liu J, Jiang W. Densification behavior

of gas and water atomized 316L stainless steel powder during selective

laser melting. Appl. Surf. Sci. 2010;256(13):4350–6.

69. Murr LE, Martinez E, Hernandez J, Collins S, Amato KN, Gaytan SM

and Shindo Patrick W. Microstructures and properties of 17-4 PH

stainless steel fabricated by selective laser melting. J. Mater. Res.

Technol. 2012;1(3):167–77.

70. Badrossamay M, Childs THC. Layer formation studies in selective laser

melting of steel powders. International Solid Freeform Fabrication

Symposium. University of Texas, Austin; 2006. p. 268–79.

71. Hauser C, Childs THC, Taylor CM, Badrossamay M, Xie J, Fox P and

Neil WO. Direct selective laser sintering of tool steel powders to high

density part: effects of laser beam width and scan strategy. International

Solid Freeform Fabrication Symposium. University of Texas, Austin;

2003. p. 644–55.

72. Kempen K, Thijs L, Yasa E, Badrossamay M, Verheecke W, Kruth J.

Process optimisation and microstructural analysis for selective laser

melting of AlSi10Mg. International Solid Freeform Fabrication


73. Vora P, Derguti F, Mumtaz K, Todd I, Hopkinson N. Investigating a

semi-solid processing technique using metal powder bed additive

manufacturing processes. International Solid Freeform Fabrication


74. Yadroitsev I, Smurov I. Selective laser melting technology: From the

single laser melted track stability to 3D parts of complex shape. Phys.

Procedia. 2010;5:551–60.

75. De Peppo GM, Palmquist A, Borchardt P, Lennerås M, Hyllner J, Snis

A,Lausmaa J, Thomsen P and Karlsson C. Free-form-fabricated

commercially pure Ti and Ti6Al4V porous scaffolds support the growth

References 167

of human embryonic stem cell-derived mesodermal progenitors. Sci.

World J. 2012;2012:1–14.

76. Yadroitsev I, Bertrand P, Smurov I. Parametric analysis of the selective

laser melting process. Appl. Surf. Sci. 2007;253(19):8064–9.

77. Mumtaz K, Hopkinson N. Selective laser melting of Inconel 625 using

pulse shaping. Rapid Prototyp. J. 2010;16(4):248–57.

78. Ng CC, Savalani M, Man HC. Fabrication of magnesium using selective

laser melting technique. Rapid Prototyp. J. 2011;17(6):479–90.

79. Shishkovsky I, Yadroitsev I, Smurov I. Direct selective laser melting of

nitinol powder. Phys. Procedia. 2012;39:447–54.

80. Vision Systems Design Magazine. 2011;16.

81. Liu B. Further Process Understanding and Prediction on Selective Laser

Melting of Stainless Steel 316L [PhD Thesis]. Loughborough

University, UK; 2013.

82. Kruth JP, Mercelis P, Froyen L, Rombouts M. Binding Mechanisms in

Selective Laser Sintering and Selective Laser Melting. Rapid Prototyp.

J. 2005;11(1):25–36.

83. Seyda V, Kaufmann N, Emmelmann C. Investigation of aging processes

of Ti-6Al-4 V powder material in laser melting. Phys. Procedia.

2012;39:425–31.

84. Khan M. Selective laser melting (SLM) of gold (Au) [PhD Thesis].

Loughborough University, UK.; 2010. p. 1–152.

85. Marcu T, Todea M, Gligor I, Berce P, Popa C. Effect of surface

conditioning on the flowability of Ti6Al7Nb powder for selective laser

melting applications. Appl. Surf. Sci.; 2012 Jan;258(7):3276–82.

86. Mumtaz K, Hopkinson N. Selective laser melting of thin wall parts

using pulse shaping. J. Mater. Process. Technol. 2010;210(2):279–87.

87. Alkahari MR, Furumoto T, Ueda T, Hosokawa A, Tanaka R, Aziz A

and Sanusi M. Thermal Conductivity of Metal Powder and Consolidated

Material Fabricated via Selective Laser Melting. Key Eng. Mater. 2012

Nov;523-524:244–9.

88. Simonelli M, Tse Y, Tuck C. Further Understanding of Ti-6Al-4V

Selective Laser Melting Using Texture Analysis. International Solid

Freeform Fabrication Symposium. University of Texas, Austin; 2012. p.

480–9.

References 168

89. Shiomi M, Osakada K, Nakamura K, Yamashita K, Abe F. Residual

stress within metallic model made by selective laser melting process.

CIRP Ann. - Manuf. Technol. 2004;53(1):195–8.

90. Neikov O d, Naboychenko SS, Dowson G. Atomisation and granulation.

In: Neikov O, Naboychenko SS, Mourachova I, Gopienko V, Frishberg

I, Lotsko D, editors. Handb. non-ferrous Met. Technol. Appl. First edit.

Elsevier Ltd; Amsterdam; 2009. p. 122–5.

91. Rombouts M, Kruth JP, Froyen L, Mercelis P. Fundamentals of

selective laser melting of alloyed steel powders. CIRP Ann. - Manuf.

Technol. 2006;55(1):187–92.

92. Morgan R, Sutcliffe CJ, O’Neill W. Density analysis of direct metal

laser re-melted 316L stainless steel cubic primitives. J. Mater. Sci.

2004;39(4):1195–205.

93. Yasa E, Kruth J. Application of laser re-melting on selective laser

melting parts. Adv. Prod. Eng. Manag. 2011;6(4):259–70.

94. Yadroitsev I, Gusarov A, Yadroitsava I, Smurov I. Single track

formation in selective laser melting of metal powders. J. Mater. Process.

Technol. 2010;210(12):1624–31.

95. Vandenbroucke B, Kruth J-P. Selective laser melting of biocompatible

metals for rapid manufacturing of medical parts. Rapid Prototyp. J.

2007;13(4):196–203.

96. Tolosa I, Garciandía F, Zubiri F, Zapirain F, Esnaola A. Study of

mechanical properties of AISI 316 stainless steel processed by

“selective laser melting”, following different manufacturing strategies.

Int. J. Adv. Manuf. Technol. 2010;51(5-8):639–47.

97. Craeghs T, Bechmann F, Kruth J. Feedback control of selective laser

melting using optical sensors. Phys. Procedia. 2010;00:1–8.

98. Lott P, Schleifenbaum H, Meiners W, Wissenbach K, Hinke C,

Bültmann J. Design of an optical system for the in-situ process

monitoring of selective laser melting (SLM). Phys. Procedia.

2011;12:683–90.

99. Gusarov A, Yadroitsev I, Bertrand P, Smurov I. Heat transfer modelling

and stability analysis of selective laser melting. Appl. Surf. Sci.

2007;254(4):975–9.

100. Gibson I. Advanced Manufacturing Technology for Medical

Applications. Gibson I, editor. Sussex, England: John Wiley & Sons

Ltd.; 2005.

References 169

101. Huang SH, Liu P, Mokasdar A, Hou L. Additive manufacturing and its

societal impact: a literature review. Int. J. Adv. Manuf. Technol.

2012;67(5-8):1191–203.

102. Bibb R, Eggbeer D, Evans P, Bocca A, Sugar A. Rapid manufacture of

custom-fitting surgical guides. Rapid Prototyp. J. 2009;15(5):346–54.

103. Wehmoller M, Warnke P, Zilian C, Eufinger H. Implant design and

production—a new approach by selective laser melting. Int. Congr. Ser.

2005 May;1281:690–5.

104. Reeves P, Tuck C, Hague R. Additive manufacturing for mass

customisation. In: Fogliatto FS., da Silveira GJC, editors. Mass

Customisation. Engineering and Managing Global Operations. First

edit. Springer Verlag; 2011. p. 275–89.

105. Ryan G, Pandit A, Apatsidis DP. Fabrication methods of porous metals

for use in orthopaedic applications. Biomaterials. 2006;27(13):2651–70.

106. Pattanayak DK, Fukuda A, Matsushita T, Takemoto M, Fujibayashi S,

Sasaki K, Nishida N, Nakamura T and Kukubo T. Bioactive Ti metal

analogous to human cancellous bone : fabrication by selective laser

melting and chemical treatments. Acta Biomater. 2011;7(3):1398–406.

107. Pattanayak DK, Matsushita T, Takadama H, Fukuda A, Takemoto M,

Fujibayashi S, Sasaki K, Nishida N, Nakamura T and Kukubo T.

Fabrication of bioactive porous Ti metal with structure similar to human

cancellous bone by selective laser melting. Bioceram. Dev. Appl.

2011;2:1–3.

108. Fukuda A, Takemoto M, Saito T, Fujibayashi S, Neo M, Pattanayak

DK, Matsushita T, Sasaki K, Nishida N, Nakamura T and Kukubo T.

Osteoinduction of porous Ti implants with a channel structure fabricated

by selective laser melting. Acta Biomater. 2011;7(5):2327–36.

109. Warnke PH, Douglas T, Wollny P, Sherry E, Steiner M, Galonska S and

Sivanathan S. Rapid prototyping : porous titanium alloy scaffolds

produced by selective laser melting for bone tissue engineering. Tissue

Eng. Part C. 2009;15(2):115–24.

110. Traini T, Mangano C, Sammons RL, Mangano F, Macchi A, Piattelli A.

Direct laser metal sintering as a new approach to fabrication of an

isoelastic functionally graded material for manufacture of porous

titanium dental implants. Dent. Mater. 2008;24(11):1525–33.

111. Zieliński A, Sobieszczyk S, Seramak T, Serbiński W, Świeczko-Żurek

B, Ossowska A. Biocompatibility and bioactivity of load-bearing

metallic implants. Adv. Mater. Sci. 2010;10(4):21–31.

References 170

112. Drstvensek I, Hren NI, Strojnik T, Brajlih T, Valentan B. Applications

of rapid prototyping in cranio-maxilofacial surgery procedures. Int J

Biol Biomed Eng. 2008;2(1):29–38.

113. Figliuzzi M, Mangano F, Mangano C. A novel root analogue dental

implant using CT scan and CAD/CAM: selective laser melting

technology. Int. J. Oral Maxillofac. Surg. 2012;41(7):858–62.

114. Almeida EO, Júnior ACF, Bonfante E a, Silva NRF a, Coelho PG.

Reliability evaluation of alumina-blasted/acid-etched versus laser-

sintered dental implants. Lasers Med. Sci. 2013;28(3):851–8.

115. Castner DG, Ratner BD. Biomedical surface science: foundations to

frontiers. Surf. Sci. 2002;500(1-3):28–60.

116. Goddard J, Hotchkiss J. Polymer surface modification for the

attachment of bioactive compounds. Prog. Polym. Sci. 2007;32(7):698–

725.

117. Stanford CM. Surface modification of biomedical and dental implants

and the processes of inflammation, wound healing and bone formation.

Int. J. Mol. Sci. 2010;11(1):354–69.

118. Raynor JE, Petrie T, Fears KP, Latour R, García AJ, Collard DM.

Saccharide polymer brushes to control protein and cell adhesion to

titanium. Biomacromolecules. 2009;10(4):748–55.

119. Kim H, Choi S-H, Ryu J-J, Koh S-Y, Park J-H, Lee I-S. The

biocompatibility of SLA-treated titanium implants. Biomed. Mater.

2008;3(2):1–6.

120. Lafranzo NA. Use of self-assembled monolayers to tailor surface

properties : from lubrication to neuronal development [PhD Thesis].

Washington University; 2013.

121. Alford K. Polymers and peptides attached to terminally functionalised

self-assembled monolayers: synthesis, characterisation and molecular

architecture [Master’s Thesis]. Washington State University; 2005.

122. Chaki NK, Aslam M, Sharma J, Vijayamohanan K. Applications of self-

assembled monolayers in materials chemistry. J. Chem. Sci.

2001;113(5-6):659–70.

123. Ulman A. Formation and structure of self-assembled monolayers.

Chem. Rev. 1996;96(4):1533–54.

124. Yung CW, Fiering J, Mueller AJ, Ingber DE. Micromagnetic-

microfluidic blood cleansing device. Lab Chip. 2009;9(9):1171–7.

References 171

125. Mahony CT, Richard FA, Goshal T, Holmes JD, Morris MA.

Thermodynamics - Systems in Equilibrium and Non-Equilibrium. 1st

ed. Moreno Pirajn JC, editor. Thermodyn. Syst. Equilib. Non-

equilibrium. InTech; 2011. p. 279–306.

126. Shustak G, Domb AJ, Mandler D. Preparation and characterization of n-

alkanoic acid self-assembled monolayers adsorbed on 316L stainless

steel. Langmuir. 2004;20(18):7499–506.

127. Schreiber F. Structure and growth of self-assembling monolayers. Prog.

Surf. Sci. 2000;65(5-8):151–257.

128. Whitesides GM. The “right” size in nanobiotechnology. Nat.

Biotechnol. 2003;21(10):1161–5.

129. Mastrangelo F, Fioravanti G, Quaresima R, Raffaele V, Gherlone E.

Self-assembled monolayers (SAMs): which perspectives in implant

dentistry? J. Biomater. Nanobiotechnol. 2011;02(05):533–43.

130. Buckholtz GA, Gawalt ES. Effect of alkyl chain length on carboxylic

acid SAMs on Ti-6Al-4V. Materials (Basel). 2012;5(12):1206–18.

131. Thebault P, Taffin de Givenchy E, Guittard F, Guimon C, Geribaldi S.

Self-assembled monolayers of semi-fluorinated thiols and disulfides

with a potentially antibacterial terminal fragment on gold surfaces. Thin

Solid Films. 2008;516(8):1765–72.

132. Aronsson BO, Lausmaa J, Kasemo B. Glow discharge plasma treatment

for surface cleaning and modification of metallic biomaterials. J.

Biomed. Mater. Res. 1997;35(1):49–73.

133. Metwalli E, Haines D, Becker O, Conzone S, Pantano CG. Surface

characterizations of mono-, di-, and tri-aminosilane treated glass

substrates. J. Colloid Interface Sci. 2006;298(2):825–31.

134. Durmaz F. A modular approach to functional self-assembled

monolayers [PhD Thesis (16942)]. Swizz Federal Institute of

Technology; 2006. p. 1–118.

135. Pardo L, Boland T. A quantitative approach to studying structures and

orientation at self-assembled monolayer/fluid interfaces. J. Colloid

Interface Sci. 2003;257(1):116–20.

136. Ajami E, Aguey-zinsou K. Functionalization of electropolished titanium

surfaces with silane-based self-assembled monolayers and their

application in drug delivery. J. Colloid Interface Sci. 2012;385:258–67.

137. Mani G, Feldman MD, Oh S, Agrawal CM. Surface modification of

cobalt–chromium–tungsten–nickel alloy using octadecyltrichlorosilanes.

Appl. Surf. Sci. 2009;255(11):5961–70.

References 172

138. Raynor JE. Surface modification of titanium substrates with polymer

brushes to control cell adhesion for bioapplications [PhD Thesis].

Georgia Instritute of Technology; 2008.

139. Pellerite MJ, Dunbar TD, Boardman LD, Wood EJ. Effects of

fluorination on self-assembled monolayer formation from

alkanephosphonic acids on aluminum: kinetics and structure. J. Phys.

Chem. B. 2003;107(42):11726–36.

140. Guerrero G, Alauzun JG, Granier M, Laurencin D, Mutin PH.

Phosphonate coupling molecules for the control of surface/interface

properties and the synthesis of nanomaterials. Dalton Trans.

2013;42(35):12569–85.

141. Michel R, Lussi JW, Csucs G, Reviakine I, Danuser G, Ketterer B,

Hubbel J, Textor M and Spencer N. Selective molecular assembly

patterning: a new approach to micro- and nanochemical patterning of

surfaces for biological applications. Langmuir. 2002;18(8):3281–7.

142. Gao W, Dickinson L, Grozinger C, Frederick M, Reven L. Self-

assembled monolayers of alkylphosphonic acids on metal oxides.

Langmuir. 1996;12(26):6429–35.

143. Feng G. The synthesis and characterisation of phosphonic acids for the

surface modification study on indium tin oxide [Master’s Thesis].

Georgia Institute of Technology; 2012.

144. Mani G, Torres N, Oh S. Paclitaxel delivery from cobalt-chromium

alloy surfaces using self-assembled monolayers. Biointerphases.

2011;6(2):33–42.

145. Pawsey S, Yach K, Reven L. Self-assembly of carboxyalkylphosphonic

acids on metal oxide powders. Langmuir. 2002;18(13):5205–12.

146. Biddle M. An Introduction to SAMs: Self-Assembled Monolayers in

Organic Chemistry [Internet]. chem.wisc.edu. 2001 [cited 2011 Aug

29]. Available from:

http://www.chem.wisc.edu/areas/organic/studsemin/biddle/biddle-

abs.pdf

147. Allen MT, Diele S, Harris KD., Hegmann T, Kariuki BM, Preece JA

and Tschierske C. Intermolecular organisation of triphenylene-based

discotic mesogens by interdigitation of alkyl chain. J. Mater. Chem.

2001;11:302–11.

148. Hsu C-W, Liou H-R, Su W-F, Wang L. Self-assembled monolayers of

2-(thienyl)hexylphosphonic acid on native oxide surface of silicon

fabricated by air-liquid interface-assisted method. J Colloid Interf. Sci.

2008;324(1-2):236–9.

References 173

149. Tosatti S, Michel R, Textor M, Spencer ND. Self-assembled monolayers

of dodecyl and hydroxy-dodecyl phosphates on both smooth and rough

titanium and titanium oxide surfaces. Langmuir. 2002;3858(7):3537–48.

150. Malti W El, Laurencin D, Guerrero G, Smith ME, Mutin PH. Surface

modification of calcium carbonate with phosphonic acids. J. Mater.

Chem. 2012;22:1212–8.

151. Dadafarin H, Katic J, Metikos-Hukovic M, Omanovic S.

Functionalization of a NiTi surface with phosphonic acid self-assembled

monolayers. 219th Electrochem. Soc. Meet. The Electrochemical

Society; 2011. p. 120.

152. Schwartz J, Avaltroni MJ, Danahy MP, Silverman BM, Hanson EL,

Schwarzbauer JE, Midwood K and Gawalt E . Cell attachment and

spreading on metal implant materials. Mater. Sci. Eng. C. 2003;23:395–

400.

153. Raman A, Gawalt E. Reduction of 3T3 fibroblast adhesion on SS316L

by methyl-terminated SAMs. Mater. Sci. Eng. C. 2011;30(8):1157–61.

154. Mrksich M, Whitesides GM. Using self-assembled monolayers to

understand the interactions of man-made surfaces with proteins and

cells. Annu. Rev. Biophys. Biomol. Struct. 1996;25:55–78.

155. Variola F, Yi J-H, Richert L, Wuest JD, Rosei F, Nanci A. Tailoring the

surface properties of Ti6Al4V by controlled chemical oxidation.

Biomaterials. 2008;29(10):1285–98.

156. Chang E, Lee TM. Effect of surface chemistries and characteristics of

Ti6Al4V on the Ca and P adsorption and ion dissolution in Hank’s

ethylene diamine tetra-acetic acid solution. Biomaterials. 2002;23:2917–

25.

157. Heakal FE, Awad KA. Electrochemical corrosion and passivation

behavior of titanium and its Ti-6Al-4V alloy in low and highly

concentrated HBr solutions. Int. J. Electrochem. Sci. 2011;6:6483–502.

158. Kruth J, Vandenbroucke B, Van Vaerenbergh J, Mercelis P.

Benchmarking of different SLS/SLM processes as rapid manufacturing

techniques. International Conference on Polymer Moulds and

Innovation. Ghent, Belgium; 2005. p. 1–7.

159. Hryniewicz T, Rokicki R, Rokosz K. Magnetoelectropolished titanium

biomaterial. In: Pignatello R, editor. Biomater. Sci. Eng. First edit.

Intech; 2011. p. 227–47.

160. Szewczenko J, Walke W, Nowinska K, Marciniak J. Corrosion

resistance of Ti-6Al-4V alloy after diverse surface treatments.

Corrosion. 2010;41(5):360–71.

References 174

161. El-ghannam A, Starr L, Jones J. Laminin-5 coating enhances epithelial

cell attachment spreading and hemidesmosome assembly on Ti-6Al-4V

implant material in vitro. J. Biomed. Mater. Res. 1998;41(1):30–40.

162. Kruth J, Vandenbroucke B, Van Vaerenbergh J, Naert I. Rapid

manufacturing of dental prostheses by means of selective laser sintering

/ melting. J. Dent. Technol. 2007;(2):24–32.

163. Axelsson S. Surface characterization of titanium powders with X-ray

photoelectron spectroscopy [Diploma Thesis]. Chalmers University of

Technology, Sweden; 2012. p. 1–56.

164. Wagner C, Riggs W, Davis L, Moulder J. Handbook of X-Ray

Photoelectron Spectroscopy. Perkin-Elmer Corporation, Minnesota;

1979.

165. Olefjord I. X-ray photoelectron spectroscopy. In: Brune D, Hellborg R,

Whitlow H, Hunderi O, editors. Surf. Characterisation A users source B.

WILEY, Weinheim; 1997. p. 291–319.

166. Sundgren J, Bodo P, Lundstrom I. Auger electron spectroscopic studies

of the interface between human tissue and implants of titanium and

stainless steel. J. Colloid Interface Sci. 1986 Mar;110(1):9–20.

167. Principal axes and peak fitting uncertainities - CasaXPS Manual 2.3.16

rev 1.0. 2010 p. 1–18.

168. Henss A, Rohnke M, Knaack S, Kleine-Boymann M, Leichtweiss T,

Schmitz P, El Khassawna T, Gelinsky M, Heiss C and Janek J.

Quantification of calcium content in bone by using ToF-SIMS--a first

approach. Biointerphases. 2013 Dec;8(1):31.

169. Azéma J, Guidetti B, Dewelle J, Le Calve B, Mijatovic T, Korolyov

A,Vaysee J, Malet-Martino M, Martino R and Kiss R. 7-((4-

Substituted)piperazin-1-yl) derivatives of ciprofloxacin: synthesis and in

vitro biological evaluation as potential antitumor agents. Bioorg. Med.

Chem. 2009 Aug 1;17(15):5396–407.

170. Cormier R, Burda WN, Harrington L, Edlinger J, Kodigepalli KM,

Thomas J, Kapolka R, Roma G, Anderson BE and Turos E. Studies on

the antimicrobial properties of N-acylated ciprofloxacins. Bioorg. Med.

Chem. Lett. 2012 Oct 15;22(20):6513–20.

171. Sharma P, Jain S. Synthesis and antibacterial activity if certain novel 1-

cyclopropyl-6-fluoro-1,4-dihydro-7,4-substituted-piperazin-1-yl-4-

oxoquinoline-3-carboxylates. Acta Pharm. Sci. 2008;50(1):35–40.

172. Clayden J, Greeves N, Warren S. Organic Chemistry. second. Clayden

J, Greeves N, Warren S, editors. Oxford University Press; 2012. p. 197–

230.

References 175

173. Zupan M, Turel I, Bukovec P, White AJP, Williams DJ. Synthesis and

characterization of two novel zinc (II) complexes with Ciprofloxacin.

Croat. Chem. acta. 2001;74(1):61–74.

174. Cristina E, Cazedey L, Regina H, Salgado N. Spectrophotometric

determination of Ciprofloxacin hydrochloride in ophthalmic solution.

Adv. Anal. Chem. 2012;2(6):74–9.

175. Mäkinen TJ, Veiranto M, Lankinen P, Moritz N, Jalava J, Törmälä P

and Aro HT. In vitro and in vivo release of ciprofloxacin from

osteoconductive bone defect filler. J. Antimicrob. Chemoth.

2005;56(6):1063–8.

176. Ma X, Li L, Xu C, Wei H, Wang X, Yang X. Spectroscopy and

speciation studies on the interactions of aluminum (III) with

ciprofloxacin and β-nicotinamide adenine dinucleotide phosphate in

aqueous solutions. Molecules. 2012;17(8):9379–96.

177. Shamraiz S. Optical Spectroscopic and theoretical investigations of a

series of fluoroquinolones [Master’s Thesis]. Roskilde University; 2010.

p. 1–33.

178. Gottlieb J, Fuehner T, Dierich M, Wiesner O, Simon a R, Welte T. Are

metallic stents really safe? A long-term analysis in lung transplant

recipients. Eur Respir J. 2009;34(6):1417–22.

179. Brown DF. Detection of methicillin/oxacillin resistance in

staphylococci. J. Antimicrob. Chemother. 2001;48 Suppl 1:65–70.

180. Mantel M, Wightman JP. Influence of the surface chemistry on the

wettability of stainless steel. Surf. Interface Anal. 1994;21:595–605.

181. Dementjev AP, Graaf A De, Sanden MCM Van De, Maslakov KI. X-

Ray photoelectron spectroscopy reference data for identification of the

C3 N4 phase in carbon-nitrogen films. Diam. Relat. Mater.

2000;9:1904–7.

182. Lausmaa J. Surface spectroscopic characterization of titanium implant

materials. J. Electron Spectros. Relat. Phenomena. 1996;81(3):343–61.

183. Ask M, Lausmaa J, Kasemo B. Preparation and surface spectroscopic

characterization of oxide films on Ti6Al4V. Appl. Surf. Sci.

1989;35(3):283–301.

184. Fu Y, Du H, Zhang S, Huang W. XPS characterization of surface and

interfacial structure of sputtered TiNi films on Si substrate. Mater. Sci.

Eng. A. 2005;403(1-2):25–31.

References 176

185. Leyens C. Oxidation and protection of titanium alloys and titanium

aluminides. In: Peters M, Leyens C, editors. Titan. Titan. Alloy.

Weinheim: Wiley Verlag GmbH & Co.; 2003. p. 187–223.

186. Zreiqat H, Valenzuela SM, Nissan B Ben, Roest R, Knabe C, Radlanski

RJ, Rens H and Evans PJ. The effect of surface chemistry modification

of titanium alloy on signalling pathways in human osteoblasts.

Biomaterials. 2005;26(36):7579–86.

187. Krasicka-Cydzik E. Anodic layer formation on titanium and its alloys

for biomedical applications. In: Nurul Amin A, editor. Titan. Alloy. -

Towards acheiving enhanced properties for diversified applications. 1st

ed. Coratia: Intech; 2012. p. 175–200.

188. Velten D, Biehl V, Aubertin F, Valeske B, Possart W, Breme J.

Preparation of TiO 2 layers on cp-Ti and Ti6Al4V by thermal and

anodic oxidation and by sol-gel coating techniques and their

characterization. J. Biomed. Mater. Res. 2002;59(1):18–28.

189. Stoebner SE, Mani G. Effect of processing methods on drug release

profiles of anti-restenotic self-assembled monolayers. Appl. Surf. Sci.;

2012 Apr;258(12):5061–72.

190. Hoque E, Derose J a, Hoffmann P, Mathieu HJ, Bhushan B, Cichomski

M. Phosphonate self-assembled monolayers on aluminum surfaces. J.

Chem. Phys. 2006 May 7;124(17):174710.

191. Quiñones R, Gawalt ES. Study of the formation of self-assembled

monolayers on nitinol. Langmuir. 2007 Sep 25;23(20):10123–30.

192. Kanta a., Sedev R, Ralston J. The formation and stability of self-

assembled monolayers of octadecylphosphonic acid on titania. Colloids

Surfaces A Physicochem. Eng. Asp. 2006 Dec;291(1-3):51–8.

193. Deligianni DD, Katsala N, Ladas S, Sotiropoulou D, Amedee J,

Missirlis YF. Effect of surface roughness of the titanium alloy Ti-6Al-

4V on human bone marrow cell response and on protein adsorption.

Biomaterials. 2001;22(11):1241–51.

194. Kasemo B, Lausmaa J. Biomaterial and implant surfaces: on the role of

cleanliness, contamination, and preparation procedures. J. Biomed.

Mater. Res. 1988;22(A2 Suppl):145–58.

195. Chrzanowski W, Szewczenko J. Influence of the anodic oxidation on the

physicochemical properties of the Ti6Al4V ELI alloy. J. Mater. Process.

Technol. 2005;162-163:163–8.

196. Hanawa T, Hiromoto S, Yamamoto A, Kuroda D, Asami K. XPS

characterization of the surface oxide film of 316L stainless steel samples

References 177

that were located in Quasi-biological environments. Bioengineering.

2002;43(12):3088–92.

197. Mahapatro A, Johnson DM, Patel DN, Feldman MD, Ayon A a,

Agrawal CM. Surface modification of functional self-assembled

monolayers on 316L stainless steel via lipase catalysis. Langmuir.

2006;22(3):901–5.

198. Sowinski KM, Kays MB. Determination of ciprofloxacin concentrations

in human serum and urine by HPLC with ultraviolet and fluorescence

detection. J. Clin. Pharm. Ther. 2004;29(4):381–7.

Appendix - 1 178

APPENDIX 1

Electropolishing and Mechanical Testing of Tracheobronchial Stents

Fabricated Using Selective Laser Melting

ABSTRACT

This study explores the potential of selective laser melting (SLM) to fabricate

tracheobronchial stent and the effectiveness of electropolishing to polish this

SLM fabricated stent. The stents were fabricated in a SLM Realizer 100 using

316L stainless steel (316L SS). SLM was found to be capable of fabricating

tracheobronchial stents although it afforded a rough surface and the stents

exhibited non-elastic behaviour. The observed roughness was due to partially

sintered 316L SS particles on the stent surface and stepped profiles due to

layer-by-layer fabrication. A weight loss of 27.9% was observed on

electropolishing for two minutes and 48.6% on polishing for five minutes.

Hence, a significant decrease in the stent strut diameter was observed. An

addition of phosphorous from the electrolytic solution and a significant change

in the elemental composition of carbon and oxygen were observed on the

electropolished stent surface. Although electropolishing rendered a smooth

surface, the compressive strength of the stent reduced after electropolishing.

The reduction in compressive strength can be attributed to the removal of

material during electropolishing. However, the microstructure and micro-

hardness of the stent did not vary significantly after electropolishing.

INTRODUCTION

Tracheobronchial stents are hollow meshes or tubes in cylindrical, T or Y

shapes used to treat obstructions (stenosis) in trachea and/or bronchi. They are

made of metals or polymers or combinations of the two. Tracheobronchial

stents are typically fabricated using laser cutting, knitting, braiding or die

casting [1, 2]. The introduction of tracheobronchial stents has improved quality

of living by prolonging life expectancy; however, post-implant complications

including stent migration, fracture, bacterial colonisation, granulation tissue

formation have been experienced after tracheobronchial stenting [1-5].

Appendix - 1 179

Although a range of materials and manufacturing techniques have been trialled

and adopted, there is still not a perfect stent to treat tracheobronchial

obstruction in terms of shape and manufacturability [6].

It is envisaged that customised implant designs, due to their superior fit

compared to traditional off-the-shelf designs, could reduce post-implant

complications by preventing implant migration and fracture. The current

fabrication techniques including laser cutting, knitting and braiding do not

allow economical customisation of tracheobronchial stents. Knitting and

braiding of stents will impose limitations in the geometries that can be made.

To obtain customised tracheobronchial stents, moulds have been produced

using additive manufacturing (AM) techniques including stereolithography and

selective laser sintering [7]. However, fabricating stents from moulds will

increase the manufacturing time and cost involved, by adding an additional

step to the manufacturing process. Also fabricating complex designs using

moulds can be tedious. Hence a direct manufacturing technique that can

produce customised designs and complex structures is preferred.

Melgoza et al. [6] used a Fab@home system to directly print a

tracheobronchial stent from a computer aided design. The material used was

silicone. The Fab@Home builds objects layer by layer, using any material than

can be squeezed through a syringe and holds its shape [8]. Materials are

hardened by drying, heating, UV light, and other methods as necessary. The

work proposed by Melgoza et al., [6] concentrated more on developing an

integrated tool for making customised tracheal stent design to meet ideal stent

requirements. Since they have used Fab@home system to print their

conceptual design, the mechanical properties of the stent produced by this

equipment was not reported [6].

Although Fab@home system is capable of making customised design, it can

only process selected materials. Its ability to produce functional biomedical

parts using biocompatible metallic materials such as nitinol, 316L Stainless

Steel (316L SS), titanium and cobalt-chromium alloys have not been reported.

Also this system has limitations on the geometries that can be made due to its

relatively low resolution and small build area. Building an implant using this

Appendix - 1 180

technique with its current capabilities may not produce functional biomedical

implants.

Selective laser melting (SLM), a metal based AM technique is of particular

interest among researchers due to its ability to fabricate customised and

complex structures for biomedical applications [9]. Several implants for

maxillo-facial and other orthopaedic applications have been fabricated using

SLM and the capability of SLM to produce such implants is being studied [10-

13]. However, the ability of SLM to produce tracheobronchial stents, in

particular to process the thin struts required for these structures, has not been

studied. This work therefore explores the ability of SLM to fabricated

tracheobronchial stents possessing thin struts with a view to producing

customised versions.

Although SLM potentially has considerable advantages over conventional

methods to fabricate tracheobronchial stents, the surface quality of SLM

fabricated surfaces is rough and for stenting applications, a smooth surface is

required to prevent restenosis [14]. There are different techniques such as sand

blasting, grinding or etching to attain a good surface finish, but not all of these

techniques can be used to polish the stent surface since they can damage the

thin struts.

Electro-polishing has been recognised as a suitable technique to achieve a

superior surface finish in several biomedical applications including the

polishing of cardiovascular stents and surgical equipment [15]. Electro-

polishing is a process by which a metallic surface is smoothed by polarising it

anodically in a suitable electrolyte [14]. During electropolishing, a work piece

is immersed in a temperature controlled electrolytic bath. The work piece

serves as anode and is connected to the positive end of the power supply. The

negative terminal of the DC power supply is connected to a cathode. When

current is passed through the anode, the metal on the surface is oxidised and

dissolved in the electrolyte whereas at the cathode, reduction occurs. As a

result of this electrolytic reaction, the surface of the work piece (anode) is

polished. Although electro-polishing is currently being adopted to polish

structures built by various techniques, its application to polish SLM fabricated

Appendix - 1 181

stent structures has not been reported. Hence, in this study, the potential of

SLM to fabricate tracheobronchial stents and the ability of electropolishing to

polish SLM fabricated surface is investigated.

MATERIALS AND METHODS

Materials

Gas atomised 316L stainless steel (SS) powder with an average particle size of

31 µm (LPW Technology Ltd., UK) was used to fabricate the stents. Ortho-

phosphoric acid (85%) and glycerol used for electropolishing were purchased

from Fisher Scientific, UK. A 316L SS plate (Hydramaster, UK) 150 mm long,

40 mm wide and 2 mm thick was used as the cathode. A SLM fabricated stent

was used as the anode.

Equipment and Methods

Design and fabrication

A tracheobronchial stent of 60 mm height and 15 mm diameter with a strut

diameter of 300 µm was designed using Siemens PLM NX 7.5 software. The

design was exported to the SLM Realizer in STL file format for fabrication.

Tracheobronchial stents were fabricated layer-by-layer in a SLM Realizer 100

(MCP Tooling Technologies Ltd., UK). The SLM Realizer 100 machine

consist of a hopper attached to a wiper, an elevator that lowers the substrate to

adjust the layer thickness and a lens that focuses the laser (50 W max) to the

build area (125 mm diameter). Before SLM, the powder from which the part

was to be fabricated was spread over the build platform from the hopper to a

pre-defined layer thickness. After the layer had been spread, the laser beam

scanned and fused the powder in the areas specified by the layer of the CAD

file.

Once the scan was complete, the elevator lowered by the pre-defined layer

thickness (50 µm) for the powder to be spread for the next layer and this

process continued until the part was completed. The principle of the SLM

process has been described in detail in previous literature [16]. The parameters

used to build the stents in this work included a laser power of 50 W, scan

Appendix - 1 182

speed of 200 mm/s, layer thickness of 50 µm and laser spot size of

approximately 29 µm. After fabrication, an electrical discharge machine

(EDM) from Renishaw Plc., was used to cut the stents from the build plate and

sonicated in methanol, ethanol and deionised water.

Electropolishing

A schematic of the experimental set-up used for electropolishing is represented

in Figure 1. A 400 ml glass beaker served as the cell for electro-polishing and a

direct current (DC) rectifier (30 V and 4 amp max.) was used as the power

supply. The electrolyte composition was obtained from previous literature [14]

whereas, the electropolishing conditions were self-optimised for the stent

design.

Figure 1 Schematic representation of the self-constructed electropolishing apparatus.

The electrolyte used for this experiment composed of a mixture of ortho-

phosphoric acid (42%), glycerol (47%) and de-ionised water (11%). The stents

were polished for 1, 2, 3, 4 and 5 minutes. With the polishing time as the only

Appendix - 1 183

variable, the voltage (10-12 V), current (2.4 amp) and the temperature of the

bath (90-96 °C) used were kept constant.

Initially the stent was polished for a minute. The polished stent was cleaned

using deionised water in a sonicator for 30 minutes (twice) and then

characterised using a scanning electron microscope (SEM). Then the same

stent was again electropolished for another minute and cleaned in a sonicator

for 30 minutes (twice). Then the electropolished surface was characterised

using SEM and X-ray photoelectron spectroscopy (XPS). Similar procedure

was used until to attain a total of five minutes of electro-polishing.

Mechanical Testing

Metallographic characterisation

Metallographic characterisation was performed on the transversal section of

the stents before and after electro-polishing. The stents were ground finely

using silicon carbide (SiC) grit paper (220, 400, 600, and 1200) and finely

polished with 6µm and 1µm diamond pastes. The samples were

ultrasonically cleaned using iso-propanol and deionised water (15 minutes

each) and dried. These mechanically polished samples were then etched

using an etchant that comprised of 5 g of ferric chloride, 1 ml of nitric acid

(con.) and 1 ml of hydrochloric acid in 100ml of deionised water. A light

optical microscope (Nikon Optiphot microscope) was used to capture the

microstructure patterns (using TS View software). Microstructure patterns of

the stents before and after electro-polishing were captured using TS View

software.

Vicker’s micro-hardness test

The Vicker’s micro-hardness of the stent was determined using a Buehler

MMT3 equipment on the cross section of the stent strut before and after

electro-polishing. The load used for the test was 100 g.

Compression test

Appendix - 1 184

Compression testing was performed on an Instron 3366 equipment. The load

required to compress 75% of the stent diameter was calculated by

compressing the stent at a rate of 5 mm/min.

Surface Characterisation

Scanning Electron Microscopy (SEM)

The surface morphology of the SLM fabricated stents before and after

electropolishing was obtained using a LEO 440 SEM. The SEM was operated

at a voltage of 10 kV. In-built software (INCD) for the point distance

measurement was used to determine the strut thickness of the stents before and

after electropolishing.

X-ray photoelectron spectroscopy (XPS)

The surface chemistry of the SLM fabricated stent was probed using XPS (VG

ESCALAB Mk l). Using aluminium (Al) Kα radiation at 8 kV, high resolution

spectra of all detected elements were collected at a pass energy of 100 eV. The

relative intensity of the detected elements was plotted against binding energy

and the chemical composition of the surface was characterised.

Measurement of weight and dimensions of the stents

An Adam ADG 3000L electronic balance was used to weigh the stents before

and after electropolishing. The diameters of the stent struts were measured

from the images obtained using SEM. Five measurements were performed on

randomly selected struts from across the stent to determine the strut diameter

before and after polishing and the corresponding mean value was calculated.

The percentage weight loss and the strut diameter reduction before and after

electro-polishing were calculated.

RESULTS

Surface Morphology after SLM fabrication

Appendix - 1 185

Figure 2 Photographic image of SLM fabricated stents (a) and surface morphology of the stent

struts obtained using an SEM (b and c).

Figure 2a shows the SLM fabricated stent. The surface morphology of the

SLM fabricated stent obtained using SEM is shown in Figures 2b and 2c. From

the figure it can be observed that powder particles are partially sintered to the

stent surface. This could be a natural consequence of laser melting since the

beam area will naturally intersect some particles only partially. However,

dissipation of energy to the powder surrounding the build during the laser scan

could also be a reason. An array of segments can be observed from the figure

and these segments or stepped profiles are due to layer-by-layer fabrication.

SLM process parameters, including laser power, hatch distance and layer

thickness, have been found to influence the part’s surface morphology and

have been discussed in previous papers [17, 18].

The average measured strut diameters (using SEM images) after fabrication

were 308 ± 9.9 µm (including the partially sintered particles) whereas the

actual designed diameter was 300 µm. This increase in strut thickness was due

to the partially sintered particles on the struts and stepped profile obtained due

to the layer-by-layer fabrication.

Surface morphology after electropolishing

During electro-polishing, burrs and other projections become the areas of very

high current density and are preferentially removed [15]. Figure 3 shows the

SLM fabricated stents struts (a) as-fabricated and b) after electropolishing for 1

minute, c) 2 minutes, d) 3 minutes, e) 4 minutes and f) 5 minutes.

Appendix - 1 186

Figure 3 SLM as-fabricated stent strut (a); Stent strut after electropolishing for 1 minute (b); 2

minutes (c); 3 minutes (d); 4 minutes (e) and 5 minutes (f).

As can be observed from the Figure 3, the partially sintered particles on the

stent surface were removed after electropolishing for 2 minutes. However, the

stepped profile due to layer-by-layer manufacturing can still be observed on

the surface until the stent was polished for 5 minutes. This will increase the

stent’s surface roughness. Hence, to further smooth the surface, the stent was

further polished for an additional three minutes.

Figure 4 Strut diameter measurement using SEM. As-fabricated stent strut (a); an

electropolished stent strut for 2 minutes (b) and an electropolished stent strut for 5 minutes (c).

Table 1 Mean weight, strut diameter, weight loss and reduction in strut diameter.

Stent Weight

(mg)

Stent

Diameter

(µm)

Weight

Loss (%)

Reduction in the Stent

Strut Diameter (%)

As-

fabricated

(from 308

µm)

As-designed

(from 300

µm)

Appendix - 1 187

As-fabricated 640 ±

1.8

308 ± 2.2 - - -

Electro-polishing

(2 minutes)

459 ±

2.7

264 ± 3.08 27.9 14.2 12

Electro-polishing

(5 minutes)

327 ±

1.5

234 ± 1.4 48.6 24 22

Figure 4 shows the stent strut before and after electropolishing for 2 and 5

minutes. The strut surface can be observed to be smooth (4c) without the

stepped profile that was observed after electropolishing for 2 minutes (4b).

Although electropolishing was efficient in removing the sintered particles and

the stepped profiles, it is important to determine the change in geometry and

the amount of material removed during the process. The observed changes in

weight and strut diameter before and after electropolishing procedures are

tabulated in Table 1. Figure 5 shows a SEM micrograph of a magnified stent

strut electropolished for 5 minutes.

After electro-polishing, a significant reduction in the stent’s weight and the

strut diameter was observed. The loss in the weight of the stent was

contributed to by both the partially sintered particles to the stent surface and

the material removed from the actual part. A loss of nearly 30% of the stent’s

weight was observed after electropolishing for 2 minutes with a 14.2% loss in

their strut diameter. Although 48.6% loss of the stent’s weight was observed

after electropolishing for 5 minutes, the total reduction in the stent’s diameter

was only 24%.

Figure 5 SEM micrograph of a magnified electropolished stent strut (1.93Kx).

Appendix - 1 188

By using the relationship V1M1=V2M2 (where V1 and M1 are the volume and

mass of the stent before polishing whereas V2 and M2 are the volume and

mass of the stent after polishing), the loss of mass after electropolishing of a

stent can be expected to scale between diameter2 (for wire) since the stent has

cylindrical geometry. The obtained results nearly agreed to the relationship i.e.,

M1/M2 = R12/R2

2 for both electropolishing procedures. However, the slight

variation observed might have been due to the presence of partially sintered

(nearly spherical) particles with spherical shape on the stent struts. This shows

that most of the material removed from the stent’s surface was from the

sintered particles. However it is difficult to exactly quantify the amount of

particles sintered to the surface.

Surface Chemistry

XPS was used to probe the SLM fabricated stent before and after

electropolishing for 5 minutes to determine the change in surface chemistry.

Table 2 shows the atomic percentage of elemental distribution on the stent

surface before and after electropolishing. Medical grade 316L SS is expected

to have iron (Fe), chromium (Cr), nickel (Ni), manganese (Mn), and

molybdenum (Mo) with some traces of silicon (Si), phosphorous (P) and sulfur

(S). From the table it can be observed that there is a significant decrease in

carbon after electropolishing. This may be because the excessive carbon

present initially on the stent surface might have been from contamination [21,

22].

Table 2 Relative atomic percentage obtained using XPS

Nature of the sample Relative atomic percentage of the detected elements

C O Fe Cr Mn P N

Before electropolishing 54.4 35.7 4.4 4.3 1.2 0 0

After electropolishing 11.3 57.6 7.4 7.9 1.5 11.6 2.7

Since the concentration of C is a lot less after electropolishing compared to its

earlier concentration, the metal to oxygen ratio was calculated to determine the

change in their surface chemistry. The as-fabricated stent surface showed its

surface oxide layer to be composed of predominantly iron oxide with a

Appendix - 1 189

comparatively lesser contribution from chromium oxide. On electropolishing,

the concentration of Cr:O increased to 0.14 whereas its initial concentration

before electropolishing was 0.12. The Cr:O (0.14) ratio was observed to be

higher than the Fe:O (0.13) ratio after electropolishing showing the surface is

enriched with chromium oxide with a comparatively less contribution from

iron oxide. After electropolishing, the Mn concentration decreased relative to

its observed initial concentration.

An addition of phosphorous to the stent surface after electro-polishing was

noted and this could be from the phosphoric acid present in the electrolyte. The

used cleaning procedure used was not sufficient to remove this phosphorous

and hence, adoption of a more suitable cleaning method is required. Nitrogen

observed after electropolishing might also be due to contamination. Surface

chemical analysis showed that electropolishing enhanced the concentration of

chromium oxide in the stent surface. Although the presence of chromium oxide

may improve the corrosion resistance, leaching of chromium ions may cause

acute and/or chronic effects. is Also, the electropolished surface was

previously reported to be enriched with hydroxyl group (-OH), yielding a

highly wettable (hydrophilic) surface and the presence of stable oxide will

increase the corrosion resistance [20].

Mechanical Properties

Metallographic characterisation

Figure 6 Microstructure patterns obtained for the stents struts fabricated using SLM. a) as-

fabricated; b) after electropolishing for 5 minutes (middle of the stent strut) and c) after

electropolishing for 5 minutes exhibiting the outer surface of the stent to show that the

electropolishing did not affect the microstructure.

Appendix - 1 190

The microstructures of the struts were studied to determine the impact of

electropolishing. Figure 6a shows the microstructure pattern of the stent before

electropolishing whereas Figures 6b and 6c show the microstructure patterns

after electropolishing. All the figures show the presence of columnar and

equiaxed grains on the stent surface. These grains show different

microstructure patterns to each other because they are differently oriented. The

microstructure patterns did not show any significant difference before or after

electropolishing confirming electropolishing does not alter the microstructural

patterns.

Micro-hardness Testing

A Vickers hardness test was performed to determine the micro-hardness of the

stent struts from their microstructures. The mean hardness for the struts before

electro-polishing was 215.2 ± 3.47 HV and after electro-polishing was 220.06

± 4.16 HV. The obtained results showed that there was no significant

difference between the hardness observed for the struts before and after

polishing. This slight deviation observed in their hardness may be due to the

presence of different micro-structural orientations within the struts.

Compression Testing

A one-off compression study was performed to determine the compressive

strength of the SLM fabricated stents before and after electropolishing (Figure

7). This study revealed a decrease in the compressive strength of the stent after

electropolishing. Although this decrease is essentially due to the amount of

material removed during electropolishing, the density of the stent could also be

a factor. Since there is a high possibility for the formation of varied

microstructure and generation of different levels of porosities during the SLM

process, the density of the part may be slightly different [21].

Appendix - 1 191

Figure 7 Compression testing of the tracheobronchial stents (before and after electropolishing)

fabricated using SLM.

In compression testing, the stent fabricated by SLM was observed to deform in

a non-elastic manner. For any stenting application, the stent should be capable

of deforming elastically under any applied load [22]. The SLM fabricated

stents did not exhibit elastic property. This could also be due to the level of

porosities and micro structural patterns.

DISCUSSION

SLM was able to fabricate stents of 15 mm diameter with a strut diameter of

300µm. However, the surface rendered by this technique is not favourable for

stenting applications. The partially sintered particles on the stent surface do not

just increases the strut thickness and surface roughness, but they could also

cause some serious disorders if the stent is implanted. For example, coughing,

a reflex action that clears airways compresses stents to half of their normal size

and leads to stent deformation. Also during a cough cycle, the air from the

lungs flow out at high velocities [23, 24] to clear the airways. If the stents with

these sintered particles were implanted, frequent deformation of the stent

structure may potentially lead to the release of these partially sintered particles.

This can potentially cause acute and/or chronic disorders by blocking alveolus

(air sacs) in the lungs. Also the increased surface roughness due to the sintered

particles may cause inflammation leading to restenosis [14, 170]. Hence it is

important that the stent surface should be smooth.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 2 4 6 8 10 12

Co

mp

ressiv

e L

oad

(K

gF

)

Compressive Extension (mm)

As-fabricated

Electro-polished (5 minutes)

9N

14N

Appendix - 1 192

Electropolishing was used to polish the rough surface in the present study to

avoid any damage to the stent struts that are possible with mechanical

polishing such as sand blasting. Electropolishing rendered a smooth surface

and it reduced the weight and the strut diameter significantly. Most of the

weight loss observed could be from the removal of sintered particles, however,

the reduction in strut diameter cannot be neglected.

Previous studies reported electropolishing of cardiovascular stents that were

cut using laser from a tube. During this laser cutting, they produce tiny burrs

on the cutting zones and since during electropolishing, these burrs become the

areas of high current densities, they get preferentially eroded. Whereas with the

SLM fabricated stent, the struts have partially sintered particles with a

diameter of 5 µm and 45 µm. By the time electropolishing removes these

partially sintered particles, it also erodes materials in other areas where there is

no sintered particles.

Although electropolishing was capable of polishing the stent without

physically damaging the struts, the amount of material removed and the

reduction in the strut diameter is not desirable. Removal of more material from

the actual design can affect the mechanical properties of the stent [14]. Hence

it is important to control material removal during electropolishing. This can be

achieved by optimising the parameters (such as the current, electrolyte, bath

temperature and time) that govern the electropolishing process.

Electropolishing is suitable for surfaces with small burrs but not to polish the

as-fabricated SLM surface with partially sintered particles. However,

electropolishing can be combined with other suitable polishing technique as

the final step to polish SLM fabricated structures. Electropolishing did not alter

the microstructure and grain distribution of the stent. Vicker’s micro-hardness

did not vary significantly showing electropolishing did not affect the micro-

hardness of the stent. The observed small variation could possibly due to the

microstructural patterns.

The study on the mechanical properties of the SLM fabricated

tracheobronchial stents are not convincing due to its non-elastic behaviour. In

Appendix - 1 193

the SLM process, the powder particles are rapidly melt and cooled. This can

contribute to the varied microstructural patterns. Also, since SLM is a powder

bed process, there is a high possibility for uneven melting and entrapment of

air in the melt pool leading to porosity [21]. All these factors could have

contributed to the non-elastic property of the stent fabricated using SLM. Post

processing such as annealing can be performed to improve the mechanical

property of SLM fabricated stents.

CONCLUSION

This study showed the potential to fabricate tracheobronchial stents with a strut

diameter of 300 µm using SLM and explored the possibility to electropolish

the SLM fabricated tracheobronchial stents. This study concludes,

The surface finish of the SLM fabricated stent was rough due to the

partially sintered particles on its surface.

Electropolishing the stents for two minutes in the electrolytic bath

removed the partially sintered particles; however, on polishing for five

minutes, the stepped profiles formed due to layer-by-layer building of

the stent were removed.

Although electropolishing was capable of polishing the surface, it

removed a significant amount of material from the stent surface leading

to reduction in the weight and strut diameter of the stent.

The surface chemistry of the stents showed a significant change after

electropolishing rendering a stable and non-corrosive surface oxide

layer.

The stents fabricated using SLM with the applied process condition did

not yield highly elastic stent to meet the current requirement.

SLM produced stents require surface finishing and hence leading to

two-step process.

Further optimisation of both the SLM and electropolishing process is

required to examine the potential of SLM to fabricate customised stents

with smooth surface.

Appendix - 1 194

REFERENCES

1. Wood D. Tracheal and Bronchial Stenting. In: Grillo H, editor. Surgery

of Trachea and Bronchi. Ontario: BC Decker Inc, Hamilton, Ontario;

2004. p. 763–90.

2. Freitag L. Airway stents. Eur Respir Mon. 2010;48:190–217.

3. Ko-Pen K, Mehta A, Turner J. Flexible Bronchoscopy. Second edition.

Massachusetts, USA: Blackwell Publishing Company; 2004.

4. Walser EM. Stent placement for tracheobronchial disease. Eur. J.

Radiol. 2005 Sep;55(3):321–30.

5. Zilberman M, Eberhart RC. Drug-eluting bioresorbable stents for

various applications. Annu. Rev. Biomed. Eng. 2006;8:153–80.

6. Melgoza EL, Serenó L, Rosell A, Ciurana J. An integrated

parameterized tool for designing a customized tracheal stent. Comput.

Des. 2012;44(12):1173–81.

7. Lim CS, Eng P, Lin SC, Chua CK, Lee YT. Rapid prototyping and

tooling of custom-made tracheobronchial stents. Int. J. Adv. Manuf.

Technol. 2002;20(1):44–9.

8. Malone E, Lipson H. Fab @ Home : the personal desktop fabricator kit.

Rapid Prototyp. J. 2007;4(April):245–55.

9. Hollister SJ, Bergman TL. Biomedical applications of integrated

additive/subtractive manufacturing. World Technology Evaluation

Centre; 2004 p. 55–62.

10. Vandenbroucke B, Kruth J-P. Selective laser melting of biocompatible

metals for rapid manufacturing of medical parts. Rapid Prototyp. J.

2007;13(4):196–203.

11. Kruth J, Vandenbroucke B, Vaerenbergh J Van, Naert I. Rapid

manufacturing of dental prostheses by means of selective laser sintering

/ melting. J. Dent. Technol. 2007;(2):24–32.

12. Lindner M, Hoeges S, Meiners W, Wissenbach K, Smeets R, Telle R,

Poprawe Reinhartm and Fischer Horst. Manufacturing of individual

biodegradable bone substitute implants using selective laser melting

technique. J. Biomed. Mater. Res-A. 2011;97(4):466–71.

13. Drstvensek I, Hren NI, Strojnik T, Brajlih T, Valentan B. Applications

of rapid prototyping in cranio-maxilofacial surgery procedures. Int. J.

Biol. Biomed. Eng. 2008;2(1):29–38.

Appendix - 1 195

14. Zhao H, Humbeeck J Van, Leuven KU, Sohier J, Scheerder IDE.

Electrochemical polishing of 316L stainless steel slotted tube coronary

stents : an investigation of material removal and surface roughness.

Prog. Biomed. Res. 2003;8(2):7–9.

15. Eliaz N, Nissan O. Innovative processes for electropolishing of medical

devices made of stainless steels. J. Biomed. Mater. Res. Part A.

2007;7(9).

16. Gibson I, Rosen DW, Stucker B. Additive Manufacturing Technologies.

Boston, MA: Springer US; 2010.

17. Yadroitsev I, Smurov I. Surface morphology in selective laser melting

of metal powders. Phys. Procedia. 2011;12:264–70.

18. Kruth J, Badrossamay M, Yasa E, Deckers J, Thijs L, Van Humbeeck J.

Part and material properties in selective laser melting of metals. 16th

Int. Symp. Electromachining. 2010. p. 1–12.

19. Hanawa T, Hiromoto S, Yamamoto A, Kuroda D, Asami K. XPS

characterization of the surface oxide film of 316L stainless steel samples

that were located in Quasi-biological environments. Bioengineering.

2002;43(12):3088–92.

20. Mantel M, Wightman JP. Influence of the surface chemistry on the

wettability of stainless steel. Surf. Interface Anal. 1994;21:595–605.

21. Hautmann H, Rieger J, Huber RM, Pfeifer KJ. Elastic deformation

properties of implanted endobronchial wire stents in benign and

malignant bronchial disease: A radiographic in vivo evaluation.

Cardiovasc. Intervent. Radiol. 1999;22(2):103–8.

22. Grillo HC. Diseases and results of treatment. In: Grillo HC, editor. Surg.

trachea bronchi. London: BC Decker Inc, USA; 2004. p. 173–429.

Appendix - 2 196

APPENDIX 2

Surface Modification of Metallic Biomaterials Using Phosphonic Acid

Monolayers

AIM

To surface modify SLM as-fabricated Ti6Al4V, 316L SS and L605 Co-Cr

surfaces using 16-phosphanohexadecanoic acid monolayers.

METHODS

Ti6Al4V, 316L SS and L605 Co-Cr substrates of dimensions 10 x 10 x 3 mm

were fabricated in a selective laser melting (SLM) machine using previously

optimised parameters. 1 mmoles of 16-phosphonohexadecanoic acid (16-

PhDA) was prepared in 20 ml of THF. Before surface modification, the as-

fabricated substrates were sonicated with dichloromethane, methanol and

deionised water for 15 minutes twice and dried to remove surface

contaminants. These cleaned surface treated SLM substrates were then rinsed

with tetrahydrofuran (THF) and immediately dipped into a solution of

1mmoles of 16-PhDA in 20 ml THF. To avoid any cross contamination, the

substrates of different materials were immersed in different containers. After

24 hours, the substrates were removed and the residual solvent was allowed to

evaporate at room temperature. Without rinsing, the samples were transferred

to a normal air-convection oven maintained at 120 °C for 18 hours. Finally, the

SAM coated substrates were allowed to cool at room temperature and

sonicated in THF and deionised water for 1 minute each to remove any

physisorbed molecules.

RESULTS AND DISCUSSIONS

Figure 1 represents the rough nature of the as-fabricated SLM samples in

various magnifications. This is due to the partial sintering of particles to the

samples during SLM process. Figure 2 represents the XPS scan (P 2p region)

of the Ti6Al4V, 316L SS and L605 Co-Cr substrates before and after surface

modification using SAMs. A metal-phosphonate peak observed at 133.3 eV for

the SAM treated sample was absent in their respective control samples. This

Appendix - 2 197

confirms that adsorption of 16-PhDA monolayers to all three metal oxide

surfaces.

Figure 1 SEM micrograph of as fabricated surfaces.

Figure 2 High resolution P 2p spectra obtained using XPS for various metal surfaces before

and after surface modification using SAMs.

CONCLUSION

Adsorption of 16-PhDA monolayers on all three SLM as-fabricated surfaces

shows the potential of 16-PhDA SAMs to modify metal oxide surfaces.

Sintered particles on the SLM surfaces did not affect the monolayer formation

significantly. However, stability of these monolayers on the rough SLM

surface should be studied. Comparison of SAM attachment on both as

fabricated SLM surface and mechanically polished SLM surface may generate

knowledge on the effect of monolayers formation on a rough and a smooth

surface.

Appendix - 3 198

APPENDIX 3

Fourier Transform Infrared Spectroscopy Results

Figure 1 FTIR Spectra obtained using Perkin Elmer Spectrum 100 FTIR-ATR for two

different Ti6Al4V samples coated with Ciprofloxacin®.

Vaithilingam, Jayasheelan (2015) Additive manufacturing and...

Documents

Transcript of Vaithilingam, Jayasheelan (2015) Additive manufacturing and...